1 |
root |
1.25 |
NAME |
2 |
root |
1.48 |
AnyEvent - the DBI of event loop programming |
3 |
root |
1.2 |
|
4 |
root |
1.49 |
EV, Event, Glib, Tk, Perl, Event::Lib, Irssi, rxvt-unicode, IO::Async, |
5 |
root |
1.66 |
Qt, FLTK and POE are various supported event loops/environments. |
6 |
root |
1.2 |
|
7 |
|
|
SYNOPSIS |
8 |
root |
1.4 |
use AnyEvent; |
9 |
root |
1.2 |
|
10 |
root |
1.62 |
# if you prefer function calls, look at the AE manpage for |
11 |
root |
1.60 |
# an alternative API. |
12 |
|
|
|
13 |
|
|
# file handle or descriptor readable |
14 |
root |
1.38 |
my $w = AnyEvent->io (fh => $fh, poll => "r", cb => sub { ... }); |
15 |
root |
1.29 |
|
16 |
root |
1.38 |
# one-shot or repeating timers |
17 |
root |
1.29 |
my $w = AnyEvent->timer (after => $seconds, cb => sub { ... }); |
18 |
root |
1.63 |
my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...); |
19 |
root |
1.29 |
|
20 |
|
|
print AnyEvent->now; # prints current event loop time |
21 |
|
|
print AnyEvent->time; # think Time::HiRes::time or simply CORE::time. |
22 |
|
|
|
23 |
root |
1.38 |
# POSIX signal |
24 |
root |
1.29 |
my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... }); |
25 |
root |
1.3 |
|
26 |
root |
1.38 |
# child process exit |
27 |
root |
1.29 |
my $w = AnyEvent->child (pid => $pid, cb => sub { |
28 |
|
|
my ($pid, $status) = @_; |
29 |
root |
1.2 |
... |
30 |
|
|
}); |
31 |
|
|
|
32 |
root |
1.38 |
# called when event loop idle (if applicable) |
33 |
|
|
my $w = AnyEvent->idle (cb => sub { ... }); |
34 |
|
|
|
35 |
root |
1.16 |
my $w = AnyEvent->condvar; # stores whether a condition was flagged |
36 |
root |
1.20 |
$w->send; # wake up current and all future recv's |
37 |
|
|
$w->recv; # enters "main loop" till $condvar gets ->send |
38 |
root |
1.29 |
# use a condvar in callback mode: |
39 |
|
|
$w->cb (sub { $_[0]->recv }); |
40 |
root |
1.3 |
|
41 |
root |
1.25 |
INTRODUCTION/TUTORIAL |
42 |
|
|
This manpage is mainly a reference manual. If you are interested in a |
43 |
|
|
tutorial or some gentle introduction, have a look at the AnyEvent::Intro |
44 |
|
|
manpage. |
45 |
|
|
|
46 |
root |
1.47 |
SUPPORT |
47 |
root |
1.63 |
An FAQ document is available as AnyEvent::FAQ. |
48 |
|
|
|
49 |
|
|
There also is a mailinglist for discussing all things AnyEvent, and an |
50 |
|
|
IRC channel, too. |
51 |
root |
1.47 |
|
52 |
|
|
See the AnyEvent project page at the Schmorpforge Ta-Sa Software |
53 |
root |
1.48 |
Repository, at <http://anyevent.schmorp.de>, for more info. |
54 |
root |
1.47 |
|
55 |
root |
1.14 |
WHY YOU SHOULD USE THIS MODULE (OR NOT) |
56 |
|
|
Glib, POE, IO::Async, Event... CPAN offers event models by the dozen |
57 |
|
|
nowadays. So what is different about AnyEvent? |
58 |
|
|
|
59 |
|
|
Executive Summary: AnyEvent is *compatible*, AnyEvent is *free of |
60 |
|
|
policy* and AnyEvent is *small and efficient*. |
61 |
|
|
|
62 |
|
|
First and foremost, *AnyEvent is not an event model* itself, it only |
63 |
root |
1.28 |
interfaces to whatever event model the main program happens to use, in a |
64 |
root |
1.14 |
pragmatic way. For event models and certain classes of immortals alike, |
65 |
root |
1.16 |
the statement "there can only be one" is a bitter reality: In general, |
66 |
|
|
only one event loop can be active at the same time in a process. |
67 |
root |
1.28 |
AnyEvent cannot change this, but it can hide the differences between |
68 |
|
|
those event loops. |
69 |
root |
1.14 |
|
70 |
|
|
The goal of AnyEvent is to offer module authors the ability to do event |
71 |
|
|
programming (waiting for I/O or timer events) without subscribing to a |
72 |
|
|
religion, a way of living, and most importantly: without forcing your |
73 |
|
|
module users into the same thing by forcing them to use the same event |
74 |
|
|
model you use. |
75 |
|
|
|
76 |
root |
1.16 |
For modules like POE or IO::Async (which is a total misnomer as it is |
77 |
|
|
actually doing all I/O *synchronously*...), using them in your module is |
78 |
root |
1.63 |
like joining a cult: After you join, you are dependent on them and you |
79 |
root |
1.28 |
cannot use anything else, as they are simply incompatible to everything |
80 |
|
|
that isn't them. What's worse, all the potential users of your module |
81 |
root |
1.16 |
are *also* forced to use the same event loop you use. |
82 |
|
|
|
83 |
|
|
AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works |
84 |
|
|
fine. AnyEvent + Tk works fine etc. etc. but none of these work together |
85 |
root |
1.64 |
with the rest: POE + EV? No go. Tk + Event? No go. Again: if your module |
86 |
|
|
uses one of those, every user of your module has to use it, too. But if |
87 |
|
|
your module uses AnyEvent, it works transparently with all event models |
88 |
|
|
it supports (including stuff like IO::Async, as long as those use one of |
89 |
|
|
the supported event loops. It is easy to add new event loops to |
90 |
root |
1.63 |
AnyEvent, too, so it is future-proof). |
91 |
root |
1.14 |
|
92 |
root |
1.16 |
In addition to being free of having to use *the one and only true event |
93 |
root |
1.14 |
model*, AnyEvent also is free of bloat and policy: with POE or similar |
94 |
root |
1.22 |
modules, you get an enormous amount of code and strict rules you have to |
95 |
root |
1.63 |
follow. AnyEvent, on the other hand, is lean and to the point, by only |
96 |
|
|
offering the functionality that is necessary, in as thin as a wrapper as |
97 |
|
|
technically possible. |
98 |
root |
1.14 |
|
99 |
root |
1.24 |
Of course, AnyEvent comes with a big (and fully optional!) toolbox of |
100 |
|
|
useful functionality, such as an asynchronous DNS resolver, 100% |
101 |
|
|
non-blocking connects (even with TLS/SSL, IPv6 and on broken platforms |
102 |
|
|
such as Windows) and lots of real-world knowledge and workarounds for |
103 |
|
|
platform bugs and differences. |
104 |
|
|
|
105 |
|
|
Now, if you *do want* lots of policy (this can arguably be somewhat |
106 |
root |
1.14 |
useful) and you want to force your users to use the one and only event |
107 |
|
|
model, you should *not* use this module. |
108 |
|
|
|
109 |
root |
1.2 |
DESCRIPTION |
110 |
root |
1.63 |
AnyEvent provides a uniform interface to various event loops. This |
111 |
|
|
allows module authors to use event loop functionality without forcing |
112 |
|
|
module users to use a specific event loop implementation (since more |
113 |
|
|
than one event loop cannot coexist peacefully). |
114 |
root |
1.2 |
|
115 |
root |
1.16 |
The interface itself is vaguely similar, but not identical to the Event |
116 |
root |
1.2 |
module. |
117 |
|
|
|
118 |
root |
1.16 |
During the first call of any watcher-creation method, the module tries |
119 |
|
|
to detect the currently loaded event loop by probing whether one of the |
120 |
root |
1.65 |
following modules is already loaded: EV, AnyEvent::Loop, Event, Glib, |
121 |
|
|
Tk, Event::Lib, Qt, POE. The first one found is used. If none are |
122 |
root |
1.63 |
detected, the module tries to load the first four modules in the order |
123 |
|
|
given; but note that if EV is not available, the pure-perl |
124 |
root |
1.65 |
AnyEvent::Loop should always work, so the other two are not normally |
125 |
|
|
tried. |
126 |
root |
1.6 |
|
127 |
|
|
Because AnyEvent first checks for modules that are already loaded, |
128 |
root |
1.16 |
loading an event model explicitly before first using AnyEvent will |
129 |
root |
1.6 |
likely make that model the default. For example: |
130 |
|
|
|
131 |
|
|
use Tk; |
132 |
|
|
use AnyEvent; |
133 |
|
|
|
134 |
|
|
# .. AnyEvent will likely default to Tk |
135 |
|
|
|
136 |
root |
1.16 |
The *likely* means that, if any module loads another event model and |
137 |
root |
1.63 |
starts using it, all bets are off - this case should be very rare |
138 |
|
|
though, as very few modules hardcode event loops without announcing this |
139 |
|
|
very loudly. |
140 |
root |
1.16 |
|
141 |
root |
1.65 |
The pure-perl implementation of AnyEvent is called "AnyEvent::Loop". |
142 |
|
|
Like other event modules you can load it explicitly and enjoy the high |
143 |
|
|
availability of that event loop :) |
144 |
root |
1.6 |
|
145 |
|
|
WATCHERS |
146 |
|
|
AnyEvent has the central concept of a *watcher*, which is an object that |
147 |
|
|
stores relevant data for each kind of event you are waiting for, such as |
148 |
root |
1.22 |
the callback to call, the file handle to watch, etc. |
149 |
root |
1.6 |
|
150 |
|
|
These watchers are normal Perl objects with normal Perl lifetime. After |
151 |
|
|
creating a watcher it will immediately "watch" for events and invoke the |
152 |
root |
1.16 |
callback when the event occurs (of course, only when the event model is |
153 |
|
|
in control). |
154 |
|
|
|
155 |
root |
1.36 |
Note that callbacks must not permanently change global variables |
156 |
|
|
potentially in use by the event loop (such as $_ or $[) and that |
157 |
root |
1.63 |
callbacks must not "die". The former is good programming practice in |
158 |
root |
1.36 |
Perl and the latter stems from the fact that exception handling differs |
159 |
|
|
widely between event loops. |
160 |
|
|
|
161 |
root |
1.63 |
To disable a watcher you have to destroy it (e.g. by setting the |
162 |
root |
1.16 |
variable you store it in to "undef" or otherwise deleting all references |
163 |
|
|
to it). |
164 |
root |
1.6 |
|
165 |
|
|
All watchers are created by calling a method on the "AnyEvent" class. |
166 |
|
|
|
167 |
root |
1.16 |
Many watchers either are used with "recursion" (repeating timers for |
168 |
|
|
example), or need to refer to their watcher object in other ways. |
169 |
|
|
|
170 |
root |
1.63 |
One way to achieve that is this pattern: |
171 |
root |
1.16 |
|
172 |
root |
1.25 |
my $w; $w = AnyEvent->type (arg => value ..., cb => sub { |
173 |
|
|
# you can use $w here, for example to undef it |
174 |
|
|
undef $w; |
175 |
|
|
}); |
176 |
root |
1.16 |
|
177 |
|
|
Note that "my $w; $w =" combination. This is necessary because in Perl, |
178 |
|
|
my variables are only visible after the statement in which they are |
179 |
|
|
declared. |
180 |
|
|
|
181 |
root |
1.19 |
I/O WATCHERS |
182 |
root |
1.50 |
$w = AnyEvent->io ( |
183 |
|
|
fh => <filehandle_or_fileno>, |
184 |
|
|
poll => <"r" or "w">, |
185 |
|
|
cb => <callback>, |
186 |
|
|
); |
187 |
|
|
|
188 |
root |
1.16 |
You can create an I/O watcher by calling the "AnyEvent->io" method with |
189 |
|
|
the following mandatory key-value pairs as arguments: |
190 |
root |
1.6 |
|
191 |
root |
1.43 |
"fh" is the Perl *file handle* (or a naked file descriptor) to watch for |
192 |
root |
1.36 |
events (AnyEvent might or might not keep a reference to this file |
193 |
|
|
handle). Note that only file handles pointing to things for which |
194 |
|
|
non-blocking operation makes sense are allowed. This includes sockets, |
195 |
|
|
most character devices, pipes, fifos and so on, but not for example |
196 |
|
|
files or block devices. |
197 |
|
|
|
198 |
root |
1.16 |
"poll" must be a string that is either "r" or "w", which creates a |
199 |
root |
1.36 |
watcher waiting for "r"eadable or "w"ritable events, respectively. |
200 |
|
|
|
201 |
|
|
"cb" is the callback to invoke each time the file handle becomes ready. |
202 |
root |
1.16 |
|
203 |
root |
1.19 |
Although the callback might get passed parameters, their value and |
204 |
|
|
presence is undefined and you cannot rely on them. Portable AnyEvent |
205 |
|
|
callbacks cannot use arguments passed to I/O watcher callbacks. |
206 |
|
|
|
207 |
|
|
The I/O watcher might use the underlying file descriptor or a copy of |
208 |
|
|
it. You must not close a file handle as long as any watcher is active on |
209 |
|
|
the underlying file descriptor. |
210 |
root |
1.16 |
|
211 |
root |
1.63 |
Some event loops issue spurious readiness notifications, so you should |
212 |
root |
1.16 |
always use non-blocking calls when reading/writing from/to your file |
213 |
|
|
handles. |
214 |
root |
1.6 |
|
215 |
root |
1.28 |
Example: wait for readability of STDIN, then read a line and disable the |
216 |
|
|
watcher. |
217 |
root |
1.6 |
|
218 |
|
|
my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub { |
219 |
|
|
chomp (my $input = <STDIN>); |
220 |
|
|
warn "read: $input\n"; |
221 |
|
|
undef $w; |
222 |
|
|
}); |
223 |
|
|
|
224 |
root |
1.8 |
TIME WATCHERS |
225 |
root |
1.50 |
$w = AnyEvent->timer (after => <seconds>, cb => <callback>); |
226 |
|
|
|
227 |
|
|
$w = AnyEvent->timer ( |
228 |
|
|
after => <fractional_seconds>, |
229 |
|
|
interval => <fractional_seconds>, |
230 |
|
|
cb => <callback>, |
231 |
|
|
); |
232 |
|
|
|
233 |
root |
1.8 |
You can create a time watcher by calling the "AnyEvent->timer" method |
234 |
root |
1.6 |
with the following mandatory arguments: |
235 |
|
|
|
236 |
root |
1.16 |
"after" specifies after how many seconds (fractional values are |
237 |
root |
1.19 |
supported) the callback should be invoked. "cb" is the callback to |
238 |
|
|
invoke in that case. |
239 |
|
|
|
240 |
|
|
Although the callback might get passed parameters, their value and |
241 |
|
|
presence is undefined and you cannot rely on them. Portable AnyEvent |
242 |
|
|
callbacks cannot use arguments passed to time watcher callbacks. |
243 |
root |
1.6 |
|
244 |
root |
1.63 |
The callback will normally be invoked only once. If you specify another |
245 |
root |
1.28 |
parameter, "interval", as a strictly positive number (> 0), then the |
246 |
|
|
callback will be invoked regularly at that interval (in fractional |
247 |
|
|
seconds) after the first invocation. If "interval" is specified with a |
248 |
root |
1.63 |
false value, then it is treated as if it were not specified at all. |
249 |
root |
1.28 |
|
250 |
|
|
The callback will be rescheduled before invoking the callback, but no |
251 |
root |
1.63 |
attempt is made to avoid timer drift in most backends, so the interval |
252 |
root |
1.28 |
is only approximate. |
253 |
root |
1.6 |
|
254 |
root |
1.28 |
Example: fire an event after 7.7 seconds. |
255 |
root |
1.6 |
|
256 |
|
|
my $w = AnyEvent->timer (after => 7.7, cb => sub { |
257 |
|
|
warn "timeout\n"; |
258 |
|
|
}); |
259 |
|
|
|
260 |
|
|
# to cancel the timer: |
261 |
root |
1.13 |
undef $w; |
262 |
root |
1.6 |
|
263 |
root |
1.28 |
Example 2: fire an event after 0.5 seconds, then roughly every second. |
264 |
root |
1.16 |
|
265 |
root |
1.28 |
my $w = AnyEvent->timer (after => 0.5, interval => 1, cb => sub { |
266 |
|
|
warn "timeout\n"; |
267 |
root |
1.16 |
}; |
268 |
|
|
|
269 |
|
|
TIMING ISSUES |
270 |
|
|
There are two ways to handle timers: based on real time (relative, "fire |
271 |
|
|
in 10 seconds") and based on wallclock time (absolute, "fire at 12 |
272 |
|
|
o'clock"). |
273 |
|
|
|
274 |
|
|
While most event loops expect timers to specified in a relative way, |
275 |
|
|
they use absolute time internally. This makes a difference when your |
276 |
|
|
clock "jumps", for example, when ntp decides to set your clock backwards |
277 |
root |
1.18 |
from the wrong date of 2014-01-01 to 2008-01-01, a watcher that is |
278 |
root |
1.63 |
supposed to fire "after a second" might actually take six years to |
279 |
root |
1.18 |
finally fire. |
280 |
root |
1.16 |
|
281 |
|
|
AnyEvent cannot compensate for this. The only event loop that is |
282 |
root |
1.63 |
conscious of these issues is EV, which offers both relative (ev_timer, |
283 |
|
|
based on true relative time) and absolute (ev_periodic, based on |
284 |
|
|
wallclock time) timers. |
285 |
root |
1.16 |
|
286 |
|
|
AnyEvent always prefers relative timers, if available, matching the |
287 |
|
|
AnyEvent API. |
288 |
|
|
|
289 |
root |
1.24 |
AnyEvent has two additional methods that return the "current time": |
290 |
|
|
|
291 |
|
|
AnyEvent->time |
292 |
|
|
This returns the "current wallclock time" as a fractional number of |
293 |
|
|
seconds since the Epoch (the same thing as "time" or |
294 |
|
|
"Time::HiRes::time" return, and the result is guaranteed to be |
295 |
|
|
compatible with those). |
296 |
|
|
|
297 |
|
|
It progresses independently of any event loop processing, i.e. each |
298 |
|
|
call will check the system clock, which usually gets updated |
299 |
|
|
frequently. |
300 |
|
|
|
301 |
|
|
AnyEvent->now |
302 |
|
|
This also returns the "current wallclock time", but unlike "time", |
303 |
|
|
above, this value might change only once per event loop iteration, |
304 |
|
|
depending on the event loop (most return the same time as "time", |
305 |
|
|
above). This is the time that AnyEvent's timers get scheduled |
306 |
|
|
against. |
307 |
|
|
|
308 |
|
|
*In almost all cases (in all cases if you don't care), this is the |
309 |
|
|
function to call when you want to know the current time.* |
310 |
|
|
|
311 |
|
|
This function is also often faster then "AnyEvent->time", and thus |
312 |
|
|
the preferred method if you want some timestamp (for example, |
313 |
root |
1.63 |
AnyEvent::Handle uses this to update its activity timeouts). |
314 |
root |
1.24 |
|
315 |
|
|
The rest of this section is only of relevance if you try to be very |
316 |
root |
1.63 |
exact with your timing; you can skip it without a bad conscience. |
317 |
root |
1.24 |
|
318 |
|
|
For a practical example of when these times differ, consider |
319 |
|
|
Event::Lib and EV and the following set-up: |
320 |
|
|
|
321 |
root |
1.63 |
The event loop is running and has just invoked one of your callbacks |
322 |
root |
1.24 |
at time=500 (assume no other callbacks delay processing). In your |
323 |
|
|
callback, you wait a second by executing "sleep 1" (blocking the |
324 |
|
|
process for a second) and then (at time=501) you create a relative |
325 |
|
|
timer that fires after three seconds. |
326 |
|
|
|
327 |
|
|
With Event::Lib, "AnyEvent->time" and "AnyEvent->now" will both |
328 |
|
|
return 501, because that is the current time, and the timer will be |
329 |
|
|
scheduled to fire at time=504 (501 + 3). |
330 |
|
|
|
331 |
|
|
With EV, "AnyEvent->time" returns 501 (as that is the current time), |
332 |
|
|
but "AnyEvent->now" returns 500, as that is the time the last event |
333 |
|
|
processing phase started. With EV, your timer gets scheduled to run |
334 |
|
|
at time=503 (500 + 3). |
335 |
|
|
|
336 |
|
|
In one sense, Event::Lib is more exact, as it uses the current time |
337 |
|
|
regardless of any delays introduced by event processing. However, |
338 |
|
|
most callbacks do not expect large delays in processing, so this |
339 |
|
|
causes a higher drift (and a lot more system calls to get the |
340 |
|
|
current time). |
341 |
|
|
|
342 |
|
|
In another sense, EV is more exact, as your timer will be scheduled |
343 |
|
|
at the same time, regardless of how long event processing actually |
344 |
|
|
took. |
345 |
|
|
|
346 |
|
|
In either case, if you care (and in most cases, you don't), then you |
347 |
|
|
can get whatever behaviour you want with any event loop, by taking |
348 |
|
|
the difference between "AnyEvent->time" and "AnyEvent->now" into |
349 |
|
|
account. |
350 |
|
|
|
351 |
root |
1.37 |
AnyEvent->now_update |
352 |
root |
1.65 |
Some event loops (such as EV or AnyEvent::Loop) cache the current |
353 |
|
|
time for each loop iteration (see the discussion of AnyEvent->now, |
354 |
|
|
above). |
355 |
root |
1.37 |
|
356 |
|
|
When a callback runs for a long time (or when the process sleeps), |
357 |
|
|
then this "current" time will differ substantially from the real |
358 |
|
|
time, which might affect timers and time-outs. |
359 |
|
|
|
360 |
|
|
When this is the case, you can call this method, which will update |
361 |
|
|
the event loop's idea of "current time". |
362 |
|
|
|
363 |
root |
1.56 |
A typical example would be a script in a web server (e.g. |
364 |
|
|
"mod_perl") - when mod_perl executes the script, then the event loop |
365 |
|
|
will have the wrong idea about the "current time" (being potentially |
366 |
|
|
far in the past, when the script ran the last time). In that case |
367 |
|
|
you should arrange a call to "AnyEvent->now_update" each time the |
368 |
|
|
web server process wakes up again (e.g. at the start of your script, |
369 |
|
|
or in a handler). |
370 |
|
|
|
371 |
root |
1.37 |
Note that updating the time *might* cause some events to be handled. |
372 |
|
|
|
373 |
root |
1.16 |
SIGNAL WATCHERS |
374 |
root |
1.50 |
$w = AnyEvent->signal (signal => <uppercase_signal_name>, cb => <callback>); |
375 |
|
|
|
376 |
root |
1.16 |
You can watch for signals using a signal watcher, "signal" is the signal |
377 |
root |
1.28 |
*name* in uppercase and without any "SIG" prefix, "cb" is the Perl |
378 |
|
|
callback to be invoked whenever a signal occurs. |
379 |
root |
1.16 |
|
380 |
root |
1.19 |
Although the callback might get passed parameters, their value and |
381 |
|
|
presence is undefined and you cannot rely on them. Portable AnyEvent |
382 |
|
|
callbacks cannot use arguments passed to signal watcher callbacks. |
383 |
|
|
|
384 |
root |
1.22 |
Multiple signal occurrences can be clumped together into one callback |
385 |
|
|
invocation, and callback invocation will be synchronous. Synchronous |
386 |
root |
1.16 |
means that it might take a while until the signal gets handled by the |
387 |
root |
1.22 |
process, but it is guaranteed not to interrupt any other callbacks. |
388 |
root |
1.16 |
|
389 |
|
|
The main advantage of using these watchers is that you can share a |
390 |
root |
1.46 |
signal between multiple watchers, and AnyEvent will ensure that signals |
391 |
|
|
will not interrupt your program at bad times. |
392 |
root |
1.16 |
|
393 |
root |
1.46 |
This watcher might use %SIG (depending on the event loop used), so |
394 |
|
|
programs overwriting those signals directly will likely not work |
395 |
|
|
correctly. |
396 |
|
|
|
397 |
root |
1.47 |
Example: exit on SIGINT |
398 |
|
|
|
399 |
|
|
my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 }); |
400 |
|
|
|
401 |
root |
1.57 |
Restart Behaviour |
402 |
|
|
While restart behaviour is up to the event loop implementation, most |
403 |
|
|
will not restart syscalls (that includes Async::Interrupt and AnyEvent's |
404 |
|
|
pure perl implementation). |
405 |
|
|
|
406 |
|
|
Safe/Unsafe Signals |
407 |
|
|
Perl signals can be either "safe" (synchronous to opcode handling) or |
408 |
|
|
"unsafe" (asynchronous) - the former might get delayed indefinitely, the |
409 |
|
|
latter might corrupt your memory. |
410 |
|
|
|
411 |
|
|
AnyEvent signal handlers are, in addition, synchronous to the event |
412 |
|
|
loop, i.e. they will not interrupt your running perl program but will |
413 |
|
|
only be called as part of the normal event handling (just like timer, |
414 |
|
|
I/O etc. callbacks, too). |
415 |
|
|
|
416 |
root |
1.47 |
Signal Races, Delays and Workarounds |
417 |
|
|
Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support attaching |
418 |
|
|
callbacks to signals in a generic way, which is a pity, as you cannot do |
419 |
root |
1.50 |
race-free signal handling in perl, requiring C libraries for this. |
420 |
root |
1.63 |
AnyEvent will try to do its best, which means in some cases, signals |
421 |
root |
1.50 |
will be delayed. The maximum time a signal might be delayed is specified |
422 |
|
|
in $AnyEvent::MAX_SIGNAL_LATENCY (default: 10 seconds). This variable |
423 |
|
|
can be changed only before the first signal watcher is created, and |
424 |
|
|
should be left alone otherwise. This variable determines how often |
425 |
|
|
AnyEvent polls for signals (in case a wake-up was missed). Higher values |
426 |
root |
1.46 |
will cause fewer spurious wake-ups, which is better for power and CPU |
427 |
root |
1.50 |
saving. |
428 |
|
|
|
429 |
|
|
All these problems can be avoided by installing the optional |
430 |
|
|
Async::Interrupt module, which works with most event loops. It will not |
431 |
|
|
work with inherently broken event loops such as Event or Event::Lib (and |
432 |
root |
1.63 |
not with POE currently, as POE does its own workaround with one-second |
433 |
root |
1.50 |
latency). For those, you just have to suffer the delays. |
434 |
root |
1.16 |
|
435 |
|
|
CHILD PROCESS WATCHERS |
436 |
root |
1.50 |
$w = AnyEvent->child (pid => <process id>, cb => <callback>); |
437 |
|
|
|
438 |
root |
1.63 |
You can also watch for a child process exit and catch its exit status. |
439 |
root |
1.16 |
|
440 |
root |
1.63 |
The child process is specified by the "pid" argument (on some backends, |
441 |
root |
1.48 |
using 0 watches for any child process exit, on others this will croak). |
442 |
|
|
The watcher will be triggered only when the child process has finished |
443 |
|
|
and an exit status is available, not on any trace events |
444 |
|
|
(stopped/continued). |
445 |
root |
1.30 |
|
446 |
|
|
The callback will be called with the pid and exit status (as returned by |
447 |
|
|
waitpid), so unlike other watcher types, you *can* rely on child watcher |
448 |
|
|
callback arguments. |
449 |
|
|
|
450 |
|
|
This watcher type works by installing a signal handler for "SIGCHLD", |
451 |
|
|
and since it cannot be shared, nothing else should use SIGCHLD or reap |
452 |
|
|
random child processes (waiting for specific child processes, e.g. |
453 |
|
|
inside "system", is just fine). |
454 |
root |
1.19 |
|
455 |
|
|
There is a slight catch to child watchers, however: you usually start |
456 |
|
|
them *after* the child process was created, and this means the process |
457 |
|
|
could have exited already (and no SIGCHLD will be sent anymore). |
458 |
|
|
|
459 |
root |
1.41 |
Not all event models handle this correctly (neither POE nor IO::Async |
460 |
|
|
do, see their AnyEvent::Impl manpages for details), but even for event |
461 |
|
|
models that *do* handle this correctly, they usually need to be loaded |
462 |
|
|
before the process exits (i.e. before you fork in the first place). |
463 |
|
|
AnyEvent's pure perl event loop handles all cases correctly regardless |
464 |
|
|
of when you start the watcher. |
465 |
root |
1.19 |
|
466 |
|
|
This means you cannot create a child watcher as the very first thing in |
467 |
|
|
an AnyEvent program, you *have* to create at least one watcher before |
468 |
|
|
you "fork" the child (alternatively, you can call "AnyEvent::detect"). |
469 |
|
|
|
470 |
root |
1.46 |
As most event loops do not support waiting for child events, they will |
471 |
root |
1.65 |
be emulated by AnyEvent in most cases, in which case the latency and |
472 |
|
|
race problems mentioned in the description of signal watchers apply. |
473 |
root |
1.46 |
|
474 |
root |
1.19 |
Example: fork a process and wait for it |
475 |
|
|
|
476 |
root |
1.25 |
my $done = AnyEvent->condvar; |
477 |
root |
1.62 |
|
478 |
|
|
my $pid = fork or exit 5; |
479 |
|
|
|
480 |
|
|
my $w = AnyEvent->child ( |
481 |
root |
1.25 |
pid => $pid, |
482 |
|
|
cb => sub { |
483 |
|
|
my ($pid, $status) = @_; |
484 |
|
|
warn "pid $pid exited with status $status"; |
485 |
|
|
$done->send; |
486 |
|
|
}, |
487 |
|
|
); |
488 |
root |
1.62 |
|
489 |
|
|
# do something else, then wait for process exit |
490 |
root |
1.25 |
$done->recv; |
491 |
root |
1.19 |
|
492 |
root |
1.38 |
IDLE WATCHERS |
493 |
root |
1.50 |
$w = AnyEvent->idle (cb => <callback>); |
494 |
|
|
|
495 |
root |
1.63 |
This will repeatedly invoke the callback after the process becomes idle, |
496 |
|
|
until either the watcher is destroyed or new events have been detected. |
497 |
root |
1.59 |
|
498 |
|
|
Idle watchers are useful when there is a need to do something, but it is |
499 |
|
|
not so important (or wise) to do it instantly. The callback will be |
500 |
|
|
invoked only when there is "nothing better to do", which is usually |
501 |
|
|
defined as "all outstanding events have been handled and no new events |
502 |
|
|
have been detected". That means that idle watchers ideally get invoked |
503 |
|
|
when the event loop has just polled for new events but none have been |
504 |
|
|
detected. Instead of blocking to wait for more events, the idle watchers |
505 |
|
|
will be invoked. |
506 |
|
|
|
507 |
|
|
Unfortunately, most event loops do not really support idle watchers |
508 |
|
|
(only EV, Event and Glib do it in a usable fashion) - for the rest, |
509 |
|
|
AnyEvent will simply call the callback "from time to time". |
510 |
root |
1.38 |
|
511 |
|
|
Example: read lines from STDIN, but only process them when the program |
512 |
|
|
is otherwise idle: |
513 |
|
|
|
514 |
|
|
my @lines; # read data |
515 |
|
|
my $idle_w; |
516 |
|
|
my $io_w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub { |
517 |
|
|
push @lines, scalar <STDIN>; |
518 |
|
|
|
519 |
|
|
# start an idle watcher, if not already done |
520 |
|
|
$idle_w ||= AnyEvent->idle (cb => sub { |
521 |
|
|
# handle only one line, when there are lines left |
522 |
|
|
if (my $line = shift @lines) { |
523 |
|
|
print "handled when idle: $line"; |
524 |
|
|
} else { |
525 |
|
|
# otherwise disable the idle watcher again |
526 |
|
|
undef $idle_w; |
527 |
|
|
} |
528 |
|
|
}); |
529 |
|
|
}); |
530 |
|
|
|
531 |
root |
1.16 |
CONDITION VARIABLES |
532 |
root |
1.50 |
$cv = AnyEvent->condvar; |
533 |
|
|
|
534 |
|
|
$cv->send (<list>); |
535 |
|
|
my @res = $cv->recv; |
536 |
|
|
|
537 |
root |
1.20 |
If you are familiar with some event loops you will know that all of them |
538 |
|
|
require you to run some blocking "loop", "run" or similar function that |
539 |
|
|
will actively watch for new events and call your callbacks. |
540 |
|
|
|
541 |
root |
1.45 |
AnyEvent is slightly different: it expects somebody else to run the |
542 |
|
|
event loop and will only block when necessary (usually when told by the |
543 |
|
|
user). |
544 |
root |
1.6 |
|
545 |
root |
1.62 |
The tool to do that is called a "condition variable", so called because |
546 |
|
|
they represent a condition that must become true. |
547 |
root |
1.6 |
|
548 |
root |
1.45 |
Now is probably a good time to look at the examples further below. |
549 |
|
|
|
550 |
root |
1.20 |
Condition variables can be created by calling the "AnyEvent->condvar" |
551 |
|
|
method, usually without arguments. The only argument pair allowed is |
552 |
|
|
"cb", which specifies a callback to be called when the condition |
553 |
root |
1.29 |
variable becomes true, with the condition variable as the first argument |
554 |
|
|
(but not the results). |
555 |
root |
1.20 |
|
556 |
root |
1.22 |
After creation, the condition variable is "false" until it becomes |
557 |
|
|
"true" by calling the "send" method (or calling the condition variable |
558 |
root |
1.23 |
as if it were a callback, read about the caveats in the description for |
559 |
|
|
the "->send" method). |
560 |
root |
1.20 |
|
561 |
root |
1.62 |
Since condition variables are the most complex part of the AnyEvent API, |
562 |
|
|
here are some different mental models of what they are - pick the ones |
563 |
|
|
you can connect to: |
564 |
|
|
|
565 |
|
|
* Condition variables are like callbacks - you can call them (and pass |
566 |
|
|
them instead of callbacks). Unlike callbacks however, you can also |
567 |
|
|
wait for them to be called. |
568 |
|
|
|
569 |
|
|
* Condition variables are signals - one side can emit or send them, |
570 |
|
|
the other side can wait for them, or install a handler that is |
571 |
|
|
called when the signal fires. |
572 |
|
|
|
573 |
|
|
* Condition variables are like "Merge Points" - points in your program |
574 |
|
|
where you merge multiple independent results/control flows into one. |
575 |
|
|
|
576 |
root |
1.63 |
* Condition variables represent a transaction - functions that start |
577 |
root |
1.62 |
some kind of transaction can return them, leaving the caller the |
578 |
|
|
choice between waiting in a blocking fashion, or setting a callback. |
579 |
|
|
|
580 |
|
|
* Condition variables represent future values, or promises to deliver |
581 |
|
|
some result, long before the result is available. |
582 |
root |
1.20 |
|
583 |
|
|
Condition variables are very useful to signal that something has |
584 |
|
|
finished, for example, if you write a module that does asynchronous http |
585 |
|
|
requests, then a condition variable would be the ideal candidate to |
586 |
|
|
signal the availability of results. The user can either act when the |
587 |
|
|
callback is called or can synchronously "->recv" for the results. |
588 |
|
|
|
589 |
|
|
You can also use them to simulate traditional event loops - for example, |
590 |
|
|
you can block your main program until an event occurs - for example, you |
591 |
|
|
could "->recv" in your main program until the user clicks the Quit |
592 |
|
|
button of your app, which would "->send" the "quit" event. |
593 |
root |
1.16 |
|
594 |
|
|
Note that condition variables recurse into the event loop - if you have |
595 |
root |
1.22 |
two pieces of code that call "->recv" in a round-robin fashion, you |
596 |
root |
1.16 |
lose. Therefore, condition variables are good to export to your caller, |
597 |
|
|
but you should avoid making a blocking wait yourself, at least in |
598 |
|
|
callbacks, as this asks for trouble. |
599 |
root |
1.14 |
|
600 |
root |
1.20 |
Condition variables are represented by hash refs in perl, and the keys |
601 |
|
|
used by AnyEvent itself are all named "_ae_XXX" to make subclassing easy |
602 |
|
|
(it is often useful to build your own transaction class on top of |
603 |
|
|
AnyEvent). To subclass, use "AnyEvent::CondVar" as base class and call |
604 |
root |
1.63 |
its "new" method in your own "new" method. |
605 |
root |
1.20 |
|
606 |
|
|
There are two "sides" to a condition variable - the "producer side" |
607 |
|
|
which eventually calls "-> send", and the "consumer side", which waits |
608 |
|
|
for the send to occur. |
609 |
root |
1.6 |
|
610 |
root |
1.22 |
Example: wait for a timer. |
611 |
root |
1.6 |
|
612 |
root |
1.60 |
# condition: "wait till the timer is fired" |
613 |
|
|
my $timer_fired = AnyEvent->condvar; |
614 |
root |
1.20 |
|
615 |
root |
1.60 |
# create the timer - we could wait for, say |
616 |
|
|
# a handle becomign ready, or even an |
617 |
|
|
# AnyEvent::HTTP request to finish, but |
618 |
root |
1.20 |
# in this case, we simply use a timer: |
619 |
|
|
my $w = AnyEvent->timer ( |
620 |
|
|
after => 1, |
621 |
root |
1.60 |
cb => sub { $timer_fired->send }, |
622 |
root |
1.20 |
); |
623 |
|
|
|
624 |
|
|
# this "blocks" (while handling events) till the callback |
625 |
root |
1.53 |
# calls ->send |
626 |
root |
1.60 |
$timer_fired->recv; |
627 |
root |
1.20 |
|
628 |
root |
1.22 |
Example: wait for a timer, but take advantage of the fact that condition |
629 |
root |
1.45 |
variables are also callable directly. |
630 |
root |
1.22 |
|
631 |
|
|
my $done = AnyEvent->condvar; |
632 |
|
|
my $delay = AnyEvent->timer (after => 5, cb => $done); |
633 |
|
|
$done->recv; |
634 |
|
|
|
635 |
root |
1.29 |
Example: Imagine an API that returns a condvar and doesn't support |
636 |
|
|
callbacks. This is how you make a synchronous call, for example from the |
637 |
|
|
main program: |
638 |
|
|
|
639 |
|
|
use AnyEvent::CouchDB; |
640 |
|
|
|
641 |
|
|
... |
642 |
|
|
|
643 |
|
|
my @info = $couchdb->info->recv; |
644 |
|
|
|
645 |
root |
1.45 |
And this is how you would just set a callback to be called whenever the |
646 |
root |
1.29 |
results are available: |
647 |
|
|
|
648 |
|
|
$couchdb->info->cb (sub { |
649 |
|
|
my @info = $_[0]->recv; |
650 |
|
|
}); |
651 |
|
|
|
652 |
root |
1.20 |
METHODS FOR PRODUCERS |
653 |
|
|
These methods should only be used by the producing side, i.e. the |
654 |
|
|
code/module that eventually sends the signal. Note that it is also the |
655 |
|
|
producer side which creates the condvar in most cases, but it isn't |
656 |
|
|
uncommon for the consumer to create it as well. |
657 |
|
|
|
658 |
|
|
$cv->send (...) |
659 |
|
|
Flag the condition as ready - a running "->recv" and all further |
660 |
|
|
calls to "recv" will (eventually) return after this method has been |
661 |
|
|
called. If nobody is waiting the send will be remembered. |
662 |
|
|
|
663 |
|
|
If a callback has been set on the condition variable, it is called |
664 |
|
|
immediately from within send. |
665 |
|
|
|
666 |
|
|
Any arguments passed to the "send" call will be returned by all |
667 |
|
|
future "->recv" calls. |
668 |
|
|
|
669 |
root |
1.22 |
Condition variables are overloaded so one can call them directly (as |
670 |
root |
1.45 |
if they were a code reference). Calling them directly is the same as |
671 |
|
|
calling "send". |
672 |
root |
1.22 |
|
673 |
root |
1.20 |
$cv->croak ($error) |
674 |
root |
1.63 |
Similar to send, but causes all calls to "->recv" to invoke |
675 |
root |
1.20 |
"Carp::croak" with the given error message/object/scalar. |
676 |
|
|
|
677 |
|
|
This can be used to signal any errors to the condition variable |
678 |
root |
1.45 |
user/consumer. Doing it this way instead of calling "croak" directly |
679 |
root |
1.63 |
delays the error detection, but has the overwhelming advantage that |
680 |
root |
1.45 |
it diagnoses the error at the place where the result is expected, |
681 |
root |
1.63 |
and not deep in some event callback with no connection to the actual |
682 |
root |
1.45 |
code causing the problem. |
683 |
root |
1.20 |
|
684 |
|
|
$cv->begin ([group callback]) |
685 |
|
|
$cv->end |
686 |
|
|
These two methods can be used to combine many transactions/events |
687 |
|
|
into one. For example, a function that pings many hosts in parallel |
688 |
|
|
might want to use a condition variable for the whole process. |
689 |
|
|
|
690 |
|
|
Every call to "->begin" will increment a counter, and every call to |
691 |
|
|
"->end" will decrement it. If the counter reaches 0 in "->end", the |
692 |
root |
1.52 |
(last) callback passed to "begin" will be executed, passing the |
693 |
|
|
condvar as first argument. That callback is *supposed* to call |
694 |
|
|
"->send", but that is not required. If no group callback was set, |
695 |
|
|
"send" will be called without any arguments. |
696 |
root |
1.20 |
|
697 |
root |
1.42 |
You can think of "$cv->send" giving you an OR condition (one call |
698 |
|
|
sends), while "$cv->begin" and "$cv->end" giving you an AND |
699 |
|
|
condition (all "begin" calls must be "end"'ed before the condvar |
700 |
|
|
sends). |
701 |
|
|
|
702 |
|
|
Let's start with a simple example: you have two I/O watchers (for |
703 |
|
|
example, STDOUT and STDERR for a program), and you want to wait for |
704 |
|
|
both streams to close before activating a condvar: |
705 |
|
|
|
706 |
|
|
my $cv = AnyEvent->condvar; |
707 |
|
|
|
708 |
|
|
$cv->begin; # first watcher |
709 |
|
|
my $w1 = AnyEvent->io (fh => $fh1, cb => sub { |
710 |
|
|
defined sysread $fh1, my $buf, 4096 |
711 |
|
|
or $cv->end; |
712 |
|
|
}); |
713 |
|
|
|
714 |
|
|
$cv->begin; # second watcher |
715 |
|
|
my $w2 = AnyEvent->io (fh => $fh2, cb => sub { |
716 |
|
|
defined sysread $fh2, my $buf, 4096 |
717 |
|
|
or $cv->end; |
718 |
|
|
}); |
719 |
|
|
|
720 |
|
|
$cv->recv; |
721 |
|
|
|
722 |
|
|
This works because for every event source (EOF on file handle), |
723 |
|
|
there is one call to "begin", so the condvar waits for all calls to |
724 |
|
|
"end" before sending. |
725 |
|
|
|
726 |
|
|
The ping example mentioned above is slightly more complicated, as |
727 |
|
|
the there are results to be passwd back, and the number of tasks |
728 |
root |
1.63 |
that are begun can potentially be zero: |
729 |
root |
1.20 |
|
730 |
|
|
my $cv = AnyEvent->condvar; |
731 |
|
|
|
732 |
|
|
my %result; |
733 |
root |
1.52 |
$cv->begin (sub { shift->send (\%result) }); |
734 |
root |
1.20 |
|
735 |
|
|
for my $host (@list_of_hosts) { |
736 |
|
|
$cv->begin; |
737 |
|
|
ping_host_then_call_callback $host, sub { |
738 |
|
|
$result{$host} = ...; |
739 |
|
|
$cv->end; |
740 |
|
|
}; |
741 |
|
|
} |
742 |
|
|
|
743 |
|
|
$cv->end; |
744 |
|
|
|
745 |
|
|
This code fragment supposedly pings a number of hosts and calls |
746 |
|
|
"send" after results for all then have have been gathered - in any |
747 |
|
|
order. To achieve this, the code issues a call to "begin" when it |
748 |
|
|
starts each ping request and calls "end" when it has received some |
749 |
|
|
result for it. Since "begin" and "end" only maintain a counter, the |
750 |
|
|
order in which results arrive is not relevant. |
751 |
|
|
|
752 |
|
|
There is an additional bracketing call to "begin" and "end" outside |
753 |
|
|
the loop, which serves two important purposes: first, it sets the |
754 |
|
|
callback to be called once the counter reaches 0, and second, it |
755 |
|
|
ensures that "send" is called even when "no" hosts are being pinged |
756 |
|
|
(the loop doesn't execute once). |
757 |
|
|
|
758 |
root |
1.42 |
This is the general pattern when you "fan out" into multiple (but |
759 |
root |
1.63 |
potentially zero) subrequests: use an outer "begin"/"end" pair to |
760 |
root |
1.42 |
set the callback and ensure "end" is called at least once, and then, |
761 |
|
|
for each subrequest you start, call "begin" and for each subrequest |
762 |
|
|
you finish, call "end". |
763 |
root |
1.20 |
|
764 |
|
|
METHODS FOR CONSUMERS |
765 |
|
|
These methods should only be used by the consuming side, i.e. the code |
766 |
|
|
awaits the condition. |
767 |
|
|
|
768 |
|
|
$cv->recv |
769 |
|
|
Wait (blocking if necessary) until the "->send" or "->croak" methods |
770 |
root |
1.63 |
have been called on $cv, while servicing other watchers normally. |
771 |
root |
1.20 |
|
772 |
|
|
You can only wait once on a condition - additional calls are valid |
773 |
|
|
but will return immediately. |
774 |
|
|
|
775 |
|
|
If an error condition has been set by calling "->croak", then this |
776 |
|
|
function will call "croak". |
777 |
|
|
|
778 |
|
|
In list context, all parameters passed to "send" will be returned, |
779 |
|
|
in scalar context only the first one will be returned. |
780 |
root |
1.6 |
|
781 |
root |
1.45 |
Note that doing a blocking wait in a callback is not supported by |
782 |
|
|
any event loop, that is, recursive invocation of a blocking "->recv" |
783 |
|
|
is not allowed, and the "recv" call will "croak" if such a condition |
784 |
|
|
is detected. This condition can be slightly loosened by using |
785 |
|
|
Coro::AnyEvent, which allows you to do a blocking "->recv" from any |
786 |
|
|
thread that doesn't run the event loop itself. |
787 |
|
|
|
788 |
root |
1.15 |
Not all event models support a blocking wait - some die in that case |
789 |
root |
1.16 |
(programs might want to do that to stay interactive), so *if you are |
790 |
root |
1.45 |
using this from a module, never require a blocking wait*. Instead, |
791 |
|
|
let the caller decide whether the call will block or not (for |
792 |
|
|
example, by coupling condition variables with some kind of request |
793 |
|
|
results and supporting callbacks so the caller knows that getting |
794 |
|
|
the result will not block, while still supporting blocking waits if |
795 |
|
|
the caller so desires). |
796 |
root |
1.20 |
|
797 |
root |
1.63 |
You can ensure that "->recv" never blocks by setting a callback and |
798 |
root |
1.20 |
only calling "->recv" from within that callback (or at a later |
799 |
|
|
time). This will work even when the event loop does not support |
800 |
|
|
blocking waits otherwise. |
801 |
|
|
|
802 |
|
|
$bool = $cv->ready |
803 |
|
|
Returns true when the condition is "true", i.e. whether "send" or |
804 |
|
|
"croak" have been called. |
805 |
|
|
|
806 |
root |
1.29 |
$cb = $cv->cb ($cb->($cv)) |
807 |
root |
1.20 |
This is a mutator function that returns the callback set and |
808 |
|
|
optionally replaces it before doing so. |
809 |
|
|
|
810 |
root |
1.63 |
The callback will be called when the condition becomes "true", i.e. |
811 |
|
|
when "send" or "croak" are called, with the only argument being the |
812 |
|
|
condition variable itself. If the condition is already true, the |
813 |
|
|
callback is called immediately when it is set. Calling "recv" inside |
814 |
|
|
the callback or at any later time is guaranteed not to block. |
815 |
root |
1.8 |
|
816 |
root |
1.43 |
SUPPORTED EVENT LOOPS/BACKENDS |
817 |
|
|
The available backend classes are (every class has its own manpage): |
818 |
root |
1.7 |
|
819 |
root |
1.43 |
Backends that are autoprobed when no other event loop can be found. |
820 |
|
|
EV is the preferred backend when no other event loop seems to be in |
821 |
root |
1.51 |
use. If EV is not installed, then AnyEvent will fall back to its own |
822 |
|
|
pure-perl implementation, which is available everywhere as it comes |
823 |
|
|
with AnyEvent itself. |
824 |
root |
1.7 |
|
825 |
root |
1.43 |
AnyEvent::Impl::EV based on EV (interface to libev, best choice). |
826 |
root |
1.65 |
AnyEvent::Impl::Perl pure-perl AnyEvent::Loop, fast and portable. |
827 |
root |
1.43 |
|
828 |
|
|
Backends that are transparently being picked up when they are used. |
829 |
root |
1.63 |
These will be used if they are already loaded when the first watcher |
830 |
|
|
is created, in which case it is assumed that the application is |
831 |
|
|
using them. This means that AnyEvent will automatically pick the |
832 |
root |
1.43 |
right backend when the main program loads an event module before |
833 |
|
|
anything starts to create watchers. Nothing special needs to be done |
834 |
|
|
by the main program. |
835 |
|
|
|
836 |
root |
1.51 |
AnyEvent::Impl::Event based on Event, very stable, few glitches. |
837 |
root |
1.43 |
AnyEvent::Impl::Glib based on Glib, slow but very stable. |
838 |
|
|
AnyEvent::Impl::Tk based on Tk, very broken. |
839 |
root |
1.18 |
AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse. |
840 |
root |
1.43 |
AnyEvent::Impl::POE based on POE, very slow, some limitations. |
841 |
root |
1.48 |
AnyEvent::Impl::Irssi used when running within irssi. |
842 |
root |
1.64 |
AnyEvent::Impl::IOAsync based on IO::Async. |
843 |
|
|
AnyEvent::Impl::Cocoa based on Cocoa::EventLoop. |
844 |
root |
1.65 |
AnyEvent::Impl::FLTK2 based on FLTK (fltk 2 binding). |
845 |
root |
1.43 |
|
846 |
|
|
Backends with special needs. |
847 |
|
|
Qt requires the Qt::Application to be instantiated first, but will |
848 |
|
|
otherwise be picked up automatically. As long as the main program |
849 |
|
|
instantiates the application before any AnyEvent watchers are |
850 |
|
|
created, everything should just work. |
851 |
|
|
|
852 |
|
|
AnyEvent::Impl::Qt based on Qt. |
853 |
|
|
|
854 |
|
|
Event loops that are indirectly supported via other backends. |
855 |
|
|
Some event loops can be supported via other modules: |
856 |
root |
1.19 |
|
857 |
root |
1.43 |
There is no direct support for WxWidgets (Wx) or Prima. |
858 |
|
|
|
859 |
|
|
WxWidgets has no support for watching file handles. However, you can |
860 |
|
|
use WxWidgets through the POE adaptor, as POE has a Wx backend that |
861 |
|
|
simply polls 20 times per second, which was considered to be too |
862 |
|
|
horrible to even consider for AnyEvent. |
863 |
|
|
|
864 |
|
|
Prima is not supported as nobody seems to be using it, but it has a |
865 |
|
|
POE backend, so it can be supported through POE. |
866 |
|
|
|
867 |
|
|
AnyEvent knows about both Prima and Wx, however, and will try to |
868 |
|
|
load POE when detecting them, in the hope that POE will pick them |
869 |
|
|
up, in which case everything will be automatic. |
870 |
|
|
|
871 |
|
|
GLOBAL VARIABLES AND FUNCTIONS |
872 |
|
|
These are not normally required to use AnyEvent, but can be useful to |
873 |
|
|
write AnyEvent extension modules. |
874 |
|
|
|
875 |
|
|
$AnyEvent::MODEL |
876 |
|
|
Contains "undef" until the first watcher is being created, before |
877 |
|
|
the backend has been autodetected. |
878 |
|
|
|
879 |
|
|
Afterwards it contains the event model that is being used, which is |
880 |
|
|
the name of the Perl class implementing the model. This class is |
881 |
root |
1.63 |
usually one of the "AnyEvent::Impl::xxx" modules, but can be any |
882 |
root |
1.43 |
other class in the case AnyEvent has been extended at runtime (e.g. |
883 |
|
|
in *rxvt-unicode* it will be "urxvt::anyevent"). |
884 |
root |
1.7 |
|
885 |
root |
1.8 |
AnyEvent::detect |
886 |
|
|
Returns $AnyEvent::MODEL, forcing autodetection of the event model |
887 |
|
|
if necessary. You should only call this function right before you |
888 |
root |
1.16 |
would have created an AnyEvent watcher anyway, that is, as late as |
889 |
root |
1.63 |
possible at runtime, and not e.g. during initialisation of your |
890 |
|
|
module. |
891 |
root |
1.43 |
|
892 |
root |
1.65 |
The effect of calling this function is as if a watcher had been |
893 |
|
|
created (specifically, actions that happen "when the first watcher |
894 |
|
|
is created" happen when calling detetc as well). |
895 |
|
|
|
896 |
root |
1.43 |
If you need to do some initialisation before AnyEvent watchers are |
897 |
|
|
created, use "post_detect". |
898 |
root |
1.8 |
|
899 |
root |
1.20 |
$guard = AnyEvent::post_detect { BLOCK } |
900 |
|
|
Arranges for the code block to be executed as soon as the event |
901 |
root |
1.63 |
model is autodetected (or immediately if that has already happened). |
902 |
root |
1.20 |
|
903 |
root |
1.43 |
The block will be executed *after* the actual backend has been |
904 |
|
|
detected ($AnyEvent::MODEL is set), but *before* any watchers have |
905 |
|
|
been created, so it is possible to e.g. patch @AnyEvent::ISA or do |
906 |
|
|
other initialisations - see the sources of AnyEvent::Strict or |
907 |
|
|
AnyEvent::AIO to see how this is used. |
908 |
|
|
|
909 |
|
|
The most common usage is to create some global watchers, without |
910 |
|
|
forcing event module detection too early, for example, AnyEvent::AIO |
911 |
|
|
creates and installs the global IO::AIO watcher in a "post_detect" |
912 |
|
|
block to avoid autodetecting the event module at load time. |
913 |
|
|
|
914 |
root |
1.20 |
If called in scalar or list context, then it creates and returns an |
915 |
|
|
object that automatically removes the callback again when it is |
916 |
root |
1.48 |
destroyed (or "undef" when the hook was immediately executed). See |
917 |
|
|
AnyEvent::AIO for a case where this is useful. |
918 |
|
|
|
919 |
|
|
Example: Create a watcher for the IO::AIO module and store it in |
920 |
root |
1.63 |
$WATCHER, but do so only do so after the event loop is initialised. |
921 |
root |
1.48 |
|
922 |
|
|
our WATCHER; |
923 |
|
|
|
924 |
|
|
my $guard = AnyEvent::post_detect { |
925 |
|
|
$WATCHER = AnyEvent->io (fh => IO::AIO::poll_fileno, poll => 'r', cb => \&IO::AIO::poll_cb); |
926 |
|
|
}; |
927 |
|
|
|
928 |
|
|
# the ||= is important in case post_detect immediately runs the block, |
929 |
|
|
# as to not clobber the newly-created watcher. assigning both watcher and |
930 |
|
|
# post_detect guard to the same variable has the advantage of users being |
931 |
|
|
# able to just C<undef $WATCHER> if the watcher causes them grief. |
932 |
|
|
|
933 |
|
|
$WATCHER ||= $guard; |
934 |
root |
1.20 |
|
935 |
|
|
@AnyEvent::post_detect |
936 |
|
|
If there are any code references in this array (you can "push" to it |
937 |
root |
1.63 |
before or after loading AnyEvent), then they will be called directly |
938 |
root |
1.20 |
after the event loop has been chosen. |
939 |
|
|
|
940 |
|
|
You should check $AnyEvent::MODEL before adding to this array, |
941 |
root |
1.43 |
though: if it is defined then the event loop has already been |
942 |
|
|
detected, and the array will be ignored. |
943 |
|
|
|
944 |
|
|
Best use "AnyEvent::post_detect { BLOCK }" when your application |
945 |
root |
1.58 |
allows it, as it takes care of these details. |
946 |
root |
1.20 |
|
947 |
root |
1.43 |
This variable is mainly useful for modules that can do something |
948 |
|
|
useful when AnyEvent is used and thus want to know when it is |
949 |
|
|
initialised, but do not need to even load it by default. This array |
950 |
|
|
provides the means to hook into AnyEvent passively, without loading |
951 |
|
|
it. |
952 |
root |
1.20 |
|
953 |
root |
1.58 |
Example: To load Coro::AnyEvent whenever Coro and AnyEvent are used |
954 |
|
|
together, you could put this into Coro (this is the actual code used |
955 |
|
|
by Coro to accomplish this): |
956 |
|
|
|
957 |
|
|
if (defined $AnyEvent::MODEL) { |
958 |
|
|
# AnyEvent already initialised, so load Coro::AnyEvent |
959 |
|
|
require Coro::AnyEvent; |
960 |
|
|
} else { |
961 |
|
|
# AnyEvent not yet initialised, so make sure to load Coro::AnyEvent |
962 |
|
|
# as soon as it is |
963 |
|
|
push @AnyEvent::post_detect, sub { require Coro::AnyEvent }; |
964 |
|
|
} |
965 |
|
|
|
966 |
root |
1.65 |
AnyEvent::postpone { BLOCK } |
967 |
|
|
Arranges for the block to be executed as soon as possible, but not |
968 |
|
|
before the call itself returns. In practise, the block will be |
969 |
|
|
executed just before the event loop polls for new events, or shortly |
970 |
|
|
afterwards. |
971 |
|
|
|
972 |
|
|
This function never returns anything (to make the "return postpone { |
973 |
|
|
... }" idiom more useful. |
974 |
|
|
|
975 |
|
|
To understand the usefulness of this function, consider a function |
976 |
|
|
that asynchronously does something for you and returns some |
977 |
|
|
transaction object or guard to let you cancel the operation. For |
978 |
|
|
example, "AnyEvent::Socket::tcp_connect": |
979 |
|
|
|
980 |
|
|
# start a conenction attempt unless one is active |
981 |
|
|
$self->{connect_guard} ||= AnyEvent::Socket::tcp_connect "www.example.net", 80, sub { |
982 |
|
|
delete $self->{connect_guard}; |
983 |
|
|
... |
984 |
|
|
}; |
985 |
|
|
|
986 |
|
|
Imagine that this function could instantly call the callback, for |
987 |
|
|
example, because it detects an obvious error such as a negative port |
988 |
|
|
number. Invoking the callback before the function returns causes |
989 |
|
|
problems however: the callback will be called and will try to delete |
990 |
|
|
the guard object. But since the function hasn't returned yet, there |
991 |
|
|
is nothing to delete. When the function eventually returns it will |
992 |
|
|
assign the guard object to "$self->{connect_guard}", where it will |
993 |
|
|
likely never be deleted, so the program thinks it is still trying to |
994 |
|
|
connect. |
995 |
|
|
|
996 |
|
|
This is where "AnyEvent::postpone" should be used. Instead of |
997 |
|
|
calling the callback directly on error: |
998 |
|
|
|
999 |
|
|
$cb->(undef), return # signal error to callback, BAD! |
1000 |
|
|
if $some_error_condition; |
1001 |
|
|
|
1002 |
|
|
It should use "postpone": |
1003 |
|
|
|
1004 |
|
|
AnyEvent::postpone { $cb->(undef) }, return # signal error to callback, later |
1005 |
|
|
if $some_error_condition; |
1006 |
|
|
|
1007 |
root |
1.66 |
AnyEvent::log $level, $msg[, @args] |
1008 |
|
|
Log the given $msg at the given $level. |
1009 |
|
|
|
1010 |
root |
1.67 |
If AnyEvent::Log is not loaded then this function makes a simple |
1011 |
|
|
test to see whether the message will be logged. If the test succeeds |
1012 |
|
|
it will load AnyEvent::Log and call "AnyEvent::Log::log" - |
1013 |
root |
1.66 |
consequently, look at the AnyEvent::Log documentation for details. |
1014 |
|
|
|
1015 |
root |
1.67 |
If the test fails it will simply return. |
1016 |
|
|
|
1017 |
root |
1.66 |
If you want to sprinkle loads of logging calls around your code, |
1018 |
|
|
consider creating a logger callback with the "AnyEvent::Log::logger" |
1019 |
root |
1.67 |
function, which can reduce typing, codesize and can reduce the |
1020 |
|
|
logging overhead enourmously. |
1021 |
root |
1.66 |
|
1022 |
root |
1.6 |
WHAT TO DO IN A MODULE |
1023 |
|
|
As a module author, you should "use AnyEvent" and call AnyEvent methods |
1024 |
|
|
freely, but you should not load a specific event module or rely on it. |
1025 |
|
|
|
1026 |
root |
1.16 |
Be careful when you create watchers in the module body - AnyEvent will |
1027 |
root |
1.6 |
decide which event module to use as soon as the first method is called, |
1028 |
|
|
so by calling AnyEvent in your module body you force the user of your |
1029 |
|
|
module to load the event module first. |
1030 |
|
|
|
1031 |
root |
1.20 |
Never call "->recv" on a condition variable unless you *know* that the |
1032 |
|
|
"->send" method has been called on it already. This is because it will |
1033 |
|
|
stall the whole program, and the whole point of using events is to stay |
1034 |
|
|
interactive. |
1035 |
root |
1.16 |
|
1036 |
root |
1.20 |
It is fine, however, to call "->recv" when the user of your module |
1037 |
root |
1.16 |
requests it (i.e. if you create a http request object ad have a method |
1038 |
root |
1.63 |
called "results" that returns the results, it may call "->recv" freely, |
1039 |
|
|
as the user of your module knows what she is doing. Always). |
1040 |
root |
1.16 |
|
1041 |
root |
1.6 |
WHAT TO DO IN THE MAIN PROGRAM |
1042 |
|
|
There will always be a single main program - the only place that should |
1043 |
|
|
dictate which event model to use. |
1044 |
|
|
|
1045 |
root |
1.63 |
If the program is not event-based, it need not do anything special, even |
1046 |
|
|
when it depends on a module that uses an AnyEvent. If the program itself |
1047 |
|
|
uses AnyEvent, but does not care which event loop is used, all it needs |
1048 |
|
|
to do is "use AnyEvent". In either case, AnyEvent will choose the best |
1049 |
|
|
available loop implementation. |
1050 |
root |
1.16 |
|
1051 |
root |
1.23 |
If the main program relies on a specific event model - for example, in |
1052 |
|
|
Gtk2 programs you have to rely on the Glib module - you should load the |
1053 |
root |
1.16 |
event module before loading AnyEvent or any module that uses it: |
1054 |
|
|
generally speaking, you should load it as early as possible. The reason |
1055 |
|
|
is that modules might create watchers when they are loaded, and AnyEvent |
1056 |
|
|
will decide on the event model to use as soon as it creates watchers, |
1057 |
root |
1.63 |
and it might choose the wrong one unless you load the correct one |
1058 |
root |
1.16 |
yourself. |
1059 |
root |
1.6 |
|
1060 |
root |
1.23 |
You can chose to use a pure-perl implementation by loading the |
1061 |
root |
1.65 |
"AnyEvent::Loop" module, which gives you similar behaviour everywhere, |
1062 |
|
|
but letting AnyEvent chose the model is generally better. |
1063 |
root |
1.23 |
|
1064 |
|
|
MAINLOOP EMULATION |
1065 |
|
|
Sometimes (often for short test scripts, or even standalone programs who |
1066 |
|
|
only want to use AnyEvent), you do not want to run a specific event |
1067 |
|
|
loop. |
1068 |
|
|
|
1069 |
|
|
In that case, you can use a condition variable like this: |
1070 |
|
|
|
1071 |
|
|
AnyEvent->condvar->recv; |
1072 |
|
|
|
1073 |
|
|
This has the effect of entering the event loop and looping forever. |
1074 |
|
|
|
1075 |
|
|
Note that usually your program has some exit condition, in which case it |
1076 |
|
|
is better to use the "traditional" approach of storing a condition |
1077 |
|
|
variable somewhere, waiting for it, and sending it when the program |
1078 |
|
|
should exit cleanly. |
1079 |
root |
1.2 |
|
1080 |
root |
1.19 |
OTHER MODULES |
1081 |
|
|
The following is a non-exhaustive list of additional modules that use |
1082 |
root |
1.43 |
AnyEvent as a client and can therefore be mixed easily with other |
1083 |
|
|
AnyEvent modules and other event loops in the same program. Some of the |
1084 |
root |
1.66 |
modules come as part of AnyEvent, the others are available via CPAN (see |
1085 |
|
|
<http://search.cpan.org/search?m=module&q=anyevent%3A%3A*> for a longer |
1086 |
|
|
non-exhaustive list), and the list is heavily biased towards modules of |
1087 |
|
|
the AnyEvent author himself :) |
1088 |
root |
1.19 |
|
1089 |
|
|
AnyEvent::Util |
1090 |
root |
1.63 |
Contains various utility functions that replace often-used blocking |
1091 |
|
|
functions such as "inet_aton" with event/callback-based versions. |
1092 |
root |
1.19 |
|
1093 |
root |
1.22 |
AnyEvent::Socket |
1094 |
|
|
Provides various utility functions for (internet protocol) sockets, |
1095 |
|
|
addresses and name resolution. Also functions to create non-blocking |
1096 |
|
|
tcp connections or tcp servers, with IPv6 and SRV record support and |
1097 |
|
|
more. |
1098 |
|
|
|
1099 |
root |
1.28 |
AnyEvent::Handle |
1100 |
|
|
Provide read and write buffers, manages watchers for reads and |
1101 |
|
|
writes, supports raw and formatted I/O, I/O queued and fully |
1102 |
root |
1.63 |
transparent and non-blocking SSL/TLS (via AnyEvent::TLS). |
1103 |
root |
1.28 |
|
1104 |
root |
1.23 |
AnyEvent::DNS |
1105 |
|
|
Provides rich asynchronous DNS resolver capabilities. |
1106 |
|
|
|
1107 |
root |
1.62 |
AnyEvent::HTTP, AnyEvent::IRC, AnyEvent::XMPP, AnyEvent::GPSD, |
1108 |
|
|
AnyEvent::IGS, AnyEvent::FCP |
1109 |
|
|
Implement event-based interfaces to the protocols of the same name |
1110 |
|
|
(for the curious, IGS is the International Go Server and FCP is the |
1111 |
|
|
Freenet Client Protocol). |
1112 |
|
|
|
1113 |
root |
1.67 |
AnyEvent::AIO |
1114 |
|
|
Truly asynchronous (as opposed to non-blocking) I/O, should be in |
1115 |
|
|
the toolbox of every event programmer. AnyEvent::AIO transparently |
1116 |
|
|
fuses IO::AIO and AnyEvent together, giving AnyEvent access to |
1117 |
|
|
event-based file I/O, and much more. |
1118 |
root |
1.62 |
|
1119 |
root |
1.67 |
AnyEvent::Filesys::Notify |
1120 |
|
|
AnyEvent is good for non-blocking stuff, but it can't detect file or |
1121 |
|
|
path changes (e.g. "watch this directory for new files", "watch this |
1122 |
|
|
file for changes"). The AnyEvent::Filesys::Notify module promises to |
1123 |
|
|
do just that in a portbale fashion, supporting inotify on GNU/Linux |
1124 |
|
|
and some weird, without doubt broken, stuff on OS X to monitor |
1125 |
|
|
files. It can fall back to blocking scans at regular intervals |
1126 |
|
|
transparently on other platforms, so it's about as portable as it |
1127 |
|
|
gets. |
1128 |
|
|
|
1129 |
|
|
(I haven't used it myself, but I haven't heard anybody complaining |
1130 |
|
|
about it yet). |
1131 |
root |
1.62 |
|
1132 |
|
|
AnyEvent::DBI |
1133 |
|
|
Executes DBI requests asynchronously in a proxy process for you, |
1134 |
root |
1.63 |
notifying you in an event-based way when the operation is finished. |
1135 |
root |
1.62 |
|
1136 |
root |
1.19 |
AnyEvent::HTTPD |
1137 |
root |
1.62 |
A simple embedded webserver. |
1138 |
root |
1.19 |
|
1139 |
|
|
AnyEvent::FastPing |
1140 |
|
|
The fastest ping in the west. |
1141 |
|
|
|
1142 |
|
|
Coro |
1143 |
root |
1.67 |
Has special support for AnyEvent via Coro::AnyEvent, which allows |
1144 |
|
|
you to simply invert the flow control - don't call us, we will call |
1145 |
|
|
you: |
1146 |
|
|
|
1147 |
|
|
async { |
1148 |
|
|
Coro::AnyEvent::sleep 5; # creates a 5s timer and waits for it |
1149 |
|
|
print "5 seconds later!\n"; |
1150 |
|
|
|
1151 |
|
|
Coro::AnyEvent::readable *STDIN; # uses an I/O watcher |
1152 |
|
|
my $line = <STDIN>; # works for ttys |
1153 |
|
|
|
1154 |
|
|
AnyEvent::HTTP::http_get "url", Coro::rouse_cb; |
1155 |
|
|
my ($body, $hdr) = Coro::rouse_wait; |
1156 |
|
|
}; |
1157 |
root |
1.20 |
|
1158 |
root |
1.51 |
SIMPLIFIED AE API |
1159 |
|
|
Starting with version 5.0, AnyEvent officially supports a second, much |
1160 |
|
|
simpler, API that is designed to reduce the calling, typing and memory |
1161 |
root |
1.60 |
overhead by using function call syntax and a fixed number of parameters. |
1162 |
root |
1.51 |
|
1163 |
|
|
See the AE manpage for details. |
1164 |
|
|
|
1165 |
root |
1.30 |
ERROR AND EXCEPTION HANDLING |
1166 |
|
|
In general, AnyEvent does not do any error handling - it relies on the |
1167 |
|
|
caller to do that if required. The AnyEvent::Strict module (see also the |
1168 |
|
|
"PERL_ANYEVENT_STRICT" environment variable, below) provides strict |
1169 |
|
|
checking of all AnyEvent methods, however, which is highly useful during |
1170 |
|
|
development. |
1171 |
|
|
|
1172 |
|
|
As for exception handling (i.e. runtime errors and exceptions thrown |
1173 |
|
|
while executing a callback), this is not only highly event-loop |
1174 |
|
|
specific, but also not in any way wrapped by this module, as this is the |
1175 |
|
|
job of the main program. |
1176 |
|
|
|
1177 |
|
|
The pure perl event loop simply re-throws the exception (usually within |
1178 |
|
|
"condvar->recv"), the Event and EV modules call "$Event/EV::DIED->()", |
1179 |
|
|
Glib uses "install_exception_handler" and so on. |
1180 |
root |
1.6 |
|
1181 |
root |
1.4 |
ENVIRONMENT VARIABLES |
1182 |
root |
1.67 |
AnyEvent supports a number of environment variables that tune the |
1183 |
|
|
runtime behaviour. They are usually evaluated when AnyEvent is loaded, |
1184 |
|
|
initialised, or a submodule that uses them is loaded. Many of them also |
1185 |
|
|
cause AnyEvent to load additional modules - for example, |
1186 |
|
|
"PERL_ANYEVENT_DEBUG_WRAP" causes the AnyEvent::Debug module to be |
1187 |
|
|
loaded. |
1188 |
|
|
|
1189 |
|
|
All the environment variables documented here start with |
1190 |
|
|
"PERL_ANYEVENT_", which is what AnyEvent considers its own namespace. |
1191 |
|
|
Other modules are encouraged (but by no means required) to use |
1192 |
|
|
"PERL_ANYEVENT_SUBMODULE" if they have registered the |
1193 |
|
|
AnyEvent::Submodule namespace on CPAN, for any submodule. For example, |
1194 |
|
|
AnyEvent::HTTP could be expected to use "PERL_ANYEVENT_HTTP_PROXY" (it |
1195 |
|
|
should not access env variables starting with "AE_", see below). |
1196 |
|
|
|
1197 |
|
|
All variables can also be set via the "AE_" prefix, that is, instead of |
1198 |
|
|
setting "PERL_ANYEVENT_VERBOSE" you can also set "AE_VERBOSE". In case |
1199 |
|
|
there is a clash btween anyevent and another program that uses |
1200 |
|
|
"AE_something" you can set the corresponding "PERL_ANYEVENT_something" |
1201 |
|
|
variable to the empty string, as those variables take precedence. |
1202 |
|
|
|
1203 |
|
|
When AnyEvent is first loaded, it copies all "AE_xxx" env variables to |
1204 |
|
|
their "PERL_ANYEVENT_xxx" counterpart unless that variable already |
1205 |
|
|
exists. If taint mode is on, then AnyEvent will remove *all* environment |
1206 |
|
|
variables starting with "PERL_ANYEVENT_" from %ENV (or replace them with |
1207 |
|
|
"undef" or the empty string, if the corresaponding "AE_" variable is |
1208 |
|
|
set). |
1209 |
|
|
|
1210 |
|
|
The exact algorithm is currently: |
1211 |
|
|
|
1212 |
|
|
1. if taint mode enabled, delete all PERL_ANYEVENT_xyz variables from %ENV |
1213 |
|
|
2. copy over AE_xyz to PERL_ANYEVENT_xyz unless the latter alraedy exists |
1214 |
|
|
3. if taint mode enabled, set all PERL_ANYEVENT_xyz variables to undef. |
1215 |
|
|
|
1216 |
|
|
This ensures that child processes will not see the "AE_" variables. |
1217 |
root |
1.40 |
|
1218 |
root |
1.67 |
The following environment variables are currently known to AnyEvent: |
1219 |
root |
1.4 |
|
1220 |
root |
1.18 |
"PERL_ANYEVENT_VERBOSE" |
1221 |
root |
1.19 |
By default, AnyEvent will be completely silent except in fatal |
1222 |
|
|
conditions. You can set this environment variable to make AnyEvent |
1223 |
root |
1.67 |
more talkative. If you want to do more than just set the global |
1224 |
|
|
logging level you should have a look at "PERL_ANYEVENT_LOG", which |
1225 |
|
|
allows much more complex specifications. |
1226 |
|
|
|
1227 |
|
|
When set to 5 or higher (warn), causes AnyEvent to warn about |
1228 |
|
|
unexpected conditions, such as not being able to load the event |
1229 |
|
|
model specified by "PERL_ANYEVENT_MODEL", or a guard callback |
1230 |
|
|
throwing an exception - this is the minimum recommended level. |
1231 |
|
|
|
1232 |
|
|
When set to 7 or higher (info), cause AnyEvent to report which event |
1233 |
|
|
model it chooses. |
1234 |
|
|
|
1235 |
|
|
When set to 8 or higher (debug), then AnyEvent will report extra |
1236 |
|
|
information on which optional modules it loads and how it implements |
1237 |
|
|
certain features. |
1238 |
|
|
|
1239 |
|
|
"PERL_ANYEVENT_LOG" |
1240 |
|
|
Accepts rather complex logging specifications. For example, you |
1241 |
|
|
could log all "debug" messages of some module to stderr, warnings |
1242 |
|
|
and above to stderr, and errors and above to syslog, with: |
1243 |
|
|
|
1244 |
|
|
PERL_ANYEVENT_LOG=Some::Module=debug,+log:filter=warn,+%syslog:%syslog=error,syslog |
1245 |
|
|
|
1246 |
|
|
For the rather extensive details, see AnyEvent::Log. |
1247 |
|
|
|
1248 |
|
|
This variable is evaluated when AnyEvent (or AnyEvent::Log) is |
1249 |
|
|
loaded, so will take effect even before AnyEvent has initialised |
1250 |
|
|
itself. |
1251 |
|
|
|
1252 |
|
|
Note that specifying this environment variable causes the |
1253 |
|
|
AnyEvent::Log module to be loaded, while "PERL_ANYEVENT_VERBOSE" |
1254 |
|
|
does not, so only using the latter saves a few hundred kB of memory |
1255 |
|
|
until the first message is being logged. |
1256 |
root |
1.46 |
|
1257 |
root |
1.28 |
"PERL_ANYEVENT_STRICT" |
1258 |
|
|
AnyEvent does not do much argument checking by default, as thorough |
1259 |
|
|
argument checking is very costly. Setting this variable to a true |
1260 |
|
|
value will cause AnyEvent to load "AnyEvent::Strict" and then to |
1261 |
|
|
thoroughly check the arguments passed to most method calls. If it |
1262 |
root |
1.41 |
finds any problems, it will croak. |
1263 |
root |
1.28 |
|
1264 |
|
|
In other words, enables "strict" mode. |
1265 |
|
|
|
1266 |
root |
1.63 |
Unlike "use strict" (or its modern cousin, "use common::sense", it |
1267 |
root |
1.46 |
is definitely recommended to keep it off in production. Keeping |
1268 |
|
|
"PERL_ANYEVENT_STRICT=1" in your environment while developing |
1269 |
|
|
programs can be very useful, however. |
1270 |
root |
1.28 |
|
1271 |
root |
1.65 |
"PERL_ANYEVENT_DEBUG_SHELL" |
1272 |
|
|
If this env variable is set, then its contents will be interpreted |
1273 |
|
|
by "AnyEvent::Socket::parse_hostport" (after replacing every |
1274 |
|
|
occurance of $$ by the process pid) and an "AnyEvent::Debug::shell" |
1275 |
|
|
is bound on that port. The shell object is saved in |
1276 |
|
|
$AnyEvent::Debug::SHELL. |
1277 |
|
|
|
1278 |
root |
1.67 |
This happens when the first watcher is created. |
1279 |
root |
1.65 |
|
1280 |
|
|
For example, to bind a debug shell on a unix domain socket in |
1281 |
|
|
/tmp/debug<pid>.sock, you could use this: |
1282 |
|
|
|
1283 |
root |
1.66 |
PERL_ANYEVENT_DEBUG_SHELL=/tmp/debug\$\$.sock perlprog |
1284 |
root |
1.65 |
|
1285 |
|
|
Note that creating sockets in /tmp is very unsafe on multiuser |
1286 |
|
|
systems. |
1287 |
|
|
|
1288 |
|
|
"PERL_ANYEVENT_DEBUG_WRAP" |
1289 |
|
|
Can be set to 0, 1 or 2 and enables wrapping of all watchers for |
1290 |
|
|
debugging purposes. See "AnyEvent::Debug::wrap" for details. |
1291 |
|
|
|
1292 |
root |
1.18 |
"PERL_ANYEVENT_MODEL" |
1293 |
|
|
This can be used to specify the event model to be used by AnyEvent, |
1294 |
root |
1.65 |
before auto detection and -probing kicks in. |
1295 |
root |
1.18 |
|
1296 |
root |
1.65 |
It normally is a string consisting entirely of ASCII letters (e.g. |
1297 |
|
|
"EV" or "IOAsync"). The string "AnyEvent::Impl::" gets prepended and |
1298 |
|
|
the resulting module name is loaded and - if the load was successful |
1299 |
|
|
- used as event model backend. If it fails to load then AnyEvent |
1300 |
|
|
will proceed with auto detection and -probing. |
1301 |
|
|
|
1302 |
|
|
If the string ends with "::" instead (e.g. "AnyEvent::Impl::EV::") |
1303 |
|
|
then nothing gets prepended and the module name is used as-is (hint: |
1304 |
|
|
"::" at the end of a string designates a module name and quotes it |
1305 |
|
|
appropriately). |
1306 |
root |
1.18 |
|
1307 |
root |
1.65 |
For example, to force the pure perl model (AnyEvent::Loop::Perl) you |
1308 |
root |
1.18 |
could start your program like this: |
1309 |
|
|
|
1310 |
root |
1.25 |
PERL_ANYEVENT_MODEL=Perl perl ... |
1311 |
root |
1.4 |
|
1312 |
root |
1.22 |
"PERL_ANYEVENT_PROTOCOLS" |
1313 |
|
|
Used by both AnyEvent::DNS and AnyEvent::Socket to determine |
1314 |
|
|
preferences for IPv4 or IPv6. The default is unspecified (and might |
1315 |
|
|
change, or be the result of auto probing). |
1316 |
|
|
|
1317 |
|
|
Must be set to a comma-separated list of protocols or address |
1318 |
|
|
families, current supported: "ipv4" and "ipv6". Only protocols |
1319 |
|
|
mentioned will be used, and preference will be given to protocols |
1320 |
|
|
mentioned earlier in the list. |
1321 |
|
|
|
1322 |
|
|
This variable can effectively be used for denial-of-service attacks |
1323 |
|
|
against local programs (e.g. when setuid), although the impact is |
1324 |
root |
1.35 |
likely small, as the program has to handle conenction and other |
1325 |
|
|
failures anyways. |
1326 |
root |
1.22 |
|
1327 |
|
|
Examples: "PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6" - prefer IPv4 over |
1328 |
|
|
IPv6, but support both and try to use both. |
1329 |
|
|
"PERL_ANYEVENT_PROTOCOLS=ipv4" - only support IPv4, never try to |
1330 |
|
|
resolve or contact IPv6 addresses. |
1331 |
|
|
"PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4" support either IPv4 or IPv6, but |
1332 |
|
|
prefer IPv6 over IPv4. |
1333 |
|
|
|
1334 |
root |
1.67 |
"PERL_ANYEVENT_HOSTS" |
1335 |
|
|
This variable, if specified, overrides the /etc/hosts file used by |
1336 |
|
|
AnyEvent::Socket"::resolve_sockaddr", i.e. hosts aliases will be |
1337 |
|
|
read from that file instead. |
1338 |
|
|
|
1339 |
root |
1.22 |
"PERL_ANYEVENT_EDNS0" |
1340 |
|
|
Used by AnyEvent::DNS to decide whether to use the EDNS0 extension |
1341 |
|
|
for DNS. This extension is generally useful to reduce DNS traffic, |
1342 |
root |
1.67 |
especially when DNSSEC is involved, but some (broken) firewalls drop |
1343 |
|
|
such DNS packets, which is why it is off by default. |
1344 |
root |
1.22 |
|
1345 |
|
|
Setting this variable to 1 will cause AnyEvent::DNS to announce |
1346 |
|
|
EDNS0 in its DNS requests. |
1347 |
|
|
|
1348 |
root |
1.24 |
"PERL_ANYEVENT_MAX_FORKS" |
1349 |
|
|
The maximum number of child processes that |
1350 |
|
|
"AnyEvent::Util::fork_call" will create in parallel. |
1351 |
|
|
|
1352 |
root |
1.43 |
"PERL_ANYEVENT_MAX_OUTSTANDING_DNS" |
1353 |
|
|
The default value for the "max_outstanding" parameter for the |
1354 |
|
|
default DNS resolver - this is the maximum number of parallel DNS |
1355 |
|
|
requests that are sent to the DNS server. |
1356 |
|
|
|
1357 |
|
|
"PERL_ANYEVENT_RESOLV_CONF" |
1358 |
root |
1.67 |
The absolute path to a resolv.conf-style file to use instead of |
1359 |
|
|
/etc/resolv.conf (or the OS-specific configuration) in the default |
1360 |
|
|
resolver, or the empty string to select the default configuration. |
1361 |
root |
1.43 |
|
1362 |
|
|
"PERL_ANYEVENT_CA_FILE", "PERL_ANYEVENT_CA_PATH". |
1363 |
|
|
When neither "ca_file" nor "ca_path" was specified during |
1364 |
|
|
AnyEvent::TLS context creation, and either of these environment |
1365 |
root |
1.67 |
variables are nonempty, they will be used to specify CA certificate |
1366 |
root |
1.43 |
locations instead of a system-dependent default. |
1367 |
|
|
|
1368 |
root |
1.46 |
"PERL_ANYEVENT_AVOID_GUARD" and "PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT" |
1369 |
|
|
When these are set to 1, then the respective modules are not loaded. |
1370 |
|
|
Mostly good for testing AnyEvent itself. |
1371 |
|
|
|
1372 |
root |
1.30 |
SUPPLYING YOUR OWN EVENT MODEL INTERFACE |
1373 |
|
|
This is an advanced topic that you do not normally need to use AnyEvent |
1374 |
|
|
in a module. This section is only of use to event loop authors who want |
1375 |
|
|
to provide AnyEvent compatibility. |
1376 |
|
|
|
1377 |
|
|
If you need to support another event library which isn't directly |
1378 |
|
|
supported by AnyEvent, you can supply your own interface to it by |
1379 |
|
|
pushing, before the first watcher gets created, the package name of the |
1380 |
|
|
event module and the package name of the interface to use onto |
1381 |
|
|
@AnyEvent::REGISTRY. You can do that before and even without loading |
1382 |
|
|
AnyEvent, so it is reasonably cheap. |
1383 |
|
|
|
1384 |
|
|
Example: |
1385 |
|
|
|
1386 |
|
|
push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::]; |
1387 |
|
|
|
1388 |
|
|
This tells AnyEvent to (literally) use the "urxvt::anyevent::" |
1389 |
|
|
package/class when it finds the "urxvt" package/module is already |
1390 |
|
|
loaded. |
1391 |
|
|
|
1392 |
|
|
When AnyEvent is loaded and asked to find a suitable event model, it |
1393 |
|
|
will first check for the presence of urxvt by trying to "use" the |
1394 |
|
|
"urxvt::anyevent" module. |
1395 |
|
|
|
1396 |
|
|
The class should provide implementations for all watcher types. See |
1397 |
|
|
AnyEvent::Impl::EV (source code), AnyEvent::Impl::Glib (Source code) and |
1398 |
|
|
so on for actual examples. Use "perldoc -m AnyEvent::Impl::Glib" to see |
1399 |
|
|
the sources. |
1400 |
|
|
|
1401 |
|
|
If you don't provide "signal" and "child" watchers than AnyEvent will |
1402 |
|
|
provide suitable (hopefully) replacements. |
1403 |
|
|
|
1404 |
|
|
The above example isn't fictitious, the *rxvt-unicode* (a.k.a. urxvt) |
1405 |
|
|
terminal emulator uses the above line as-is. An interface isn't included |
1406 |
|
|
in AnyEvent because it doesn't make sense outside the embedded |
1407 |
|
|
interpreter inside *rxvt-unicode*, and it is updated and maintained as |
1408 |
|
|
part of the *rxvt-unicode* distribution. |
1409 |
|
|
|
1410 |
|
|
*rxvt-unicode* also cheats a bit by not providing blocking access to |
1411 |
|
|
condition variables: code blocking while waiting for a condition will |
1412 |
|
|
"die". This still works with most modules/usages, and blocking calls |
1413 |
|
|
must not be done in an interactive application, so it makes sense. |
1414 |
|
|
|
1415 |
root |
1.16 |
EXAMPLE PROGRAM |
1416 |
root |
1.19 |
The following program uses an I/O watcher to read data from STDIN, a |
1417 |
root |
1.16 |
timer to display a message once per second, and a condition variable to |
1418 |
|
|
quit the program when the user enters quit: |
1419 |
root |
1.2 |
|
1420 |
|
|
use AnyEvent; |
1421 |
|
|
|
1422 |
|
|
my $cv = AnyEvent->condvar; |
1423 |
|
|
|
1424 |
root |
1.16 |
my $io_watcher = AnyEvent->io ( |
1425 |
|
|
fh => \*STDIN, |
1426 |
|
|
poll => 'r', |
1427 |
|
|
cb => sub { |
1428 |
|
|
warn "io event <$_[0]>\n"; # will always output <r> |
1429 |
|
|
chomp (my $input = <STDIN>); # read a line |
1430 |
|
|
warn "read: $input\n"; # output what has been read |
1431 |
root |
1.21 |
$cv->send if $input =~ /^q/i; # quit program if /^q/i |
1432 |
root |
1.16 |
}, |
1433 |
|
|
); |
1434 |
root |
1.2 |
|
1435 |
root |
1.54 |
my $time_watcher = AnyEvent->timer (after => 1, interval => 1, cb => sub { |
1436 |
|
|
warn "timeout\n"; # print 'timeout' at most every second |
1437 |
|
|
}); |
1438 |
root |
1.2 |
|
1439 |
root |
1.21 |
$cv->recv; # wait until user enters /^q/i |
1440 |
root |
1.2 |
|
1441 |
root |
1.3 |
REAL-WORLD EXAMPLE |
1442 |
|
|
Consider the Net::FCP module. It features (among others) the following |
1443 |
|
|
API calls, which are to freenet what HTTP GET requests are to http: |
1444 |
|
|
|
1445 |
|
|
my $data = $fcp->client_get ($url); # blocks |
1446 |
|
|
|
1447 |
|
|
my $transaction = $fcp->txn_client_get ($url); # does not block |
1448 |
|
|
$transaction->cb ( sub { ... } ); # set optional result callback |
1449 |
|
|
my $data = $transaction->result; # possibly blocks |
1450 |
|
|
|
1451 |
|
|
The "client_get" method works like "LWP::Simple::get": it requests the |
1452 |
|
|
given URL and waits till the data has arrived. It is defined to be: |
1453 |
|
|
|
1454 |
|
|
sub client_get { $_[0]->txn_client_get ($_[1])->result } |
1455 |
|
|
|
1456 |
|
|
And in fact is automatically generated. This is the blocking API of |
1457 |
|
|
Net::FCP, and it works as simple as in any other, similar, module. |
1458 |
|
|
|
1459 |
|
|
More complicated is "txn_client_get": It only creates a transaction |
1460 |
|
|
(completion, result, ...) object and initiates the transaction. |
1461 |
|
|
|
1462 |
|
|
my $txn = bless { }, Net::FCP::Txn::; |
1463 |
|
|
|
1464 |
|
|
It also creates a condition variable that is used to signal the |
1465 |
|
|
completion of the request: |
1466 |
|
|
|
1467 |
|
|
$txn->{finished} = AnyAvent->condvar; |
1468 |
|
|
|
1469 |
|
|
It then creates a socket in non-blocking mode. |
1470 |
|
|
|
1471 |
|
|
socket $txn->{fh}, ...; |
1472 |
|
|
fcntl $txn->{fh}, F_SETFL, O_NONBLOCK; |
1473 |
|
|
connect $txn->{fh}, ... |
1474 |
|
|
and !$!{EWOULDBLOCK} |
1475 |
|
|
and !$!{EINPROGRESS} |
1476 |
|
|
and Carp::croak "unable to connect: $!\n"; |
1477 |
|
|
|
1478 |
root |
1.4 |
Then it creates a write-watcher which gets called whenever an error |
1479 |
root |
1.3 |
occurs or the connection succeeds: |
1480 |
|
|
|
1481 |
|
|
$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'w', cb => sub { $txn->fh_ready_w }); |
1482 |
|
|
|
1483 |
|
|
And returns this transaction object. The "fh_ready_w" callback gets |
1484 |
|
|
called as soon as the event loop detects that the socket is ready for |
1485 |
|
|
writing. |
1486 |
|
|
|
1487 |
|
|
The "fh_ready_w" method makes the socket blocking again, writes the |
1488 |
|
|
request data and replaces the watcher by a read watcher (waiting for |
1489 |
|
|
reply data). The actual code is more complicated, but that doesn't |
1490 |
|
|
matter for this example: |
1491 |
|
|
|
1492 |
|
|
fcntl $txn->{fh}, F_SETFL, 0; |
1493 |
|
|
syswrite $txn->{fh}, $txn->{request} |
1494 |
|
|
or die "connection or write error"; |
1495 |
|
|
$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r }); |
1496 |
|
|
|
1497 |
|
|
Again, "fh_ready_r" waits till all data has arrived, and then stores the |
1498 |
root |
1.22 |
result and signals any possible waiters that the request has finished: |
1499 |
root |
1.3 |
|
1500 |
|
|
sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf}; |
1501 |
|
|
|
1502 |
|
|
if (end-of-file or data complete) { |
1503 |
|
|
$txn->{result} = $txn->{buf}; |
1504 |
root |
1.21 |
$txn->{finished}->send; |
1505 |
root |
1.4 |
$txb->{cb}->($txn) of $txn->{cb}; # also call callback |
1506 |
root |
1.3 |
} |
1507 |
|
|
|
1508 |
|
|
The "result" method, finally, just waits for the finished signal (if the |
1509 |
|
|
request was already finished, it doesn't wait, of course, and returns |
1510 |
|
|
the data: |
1511 |
|
|
|
1512 |
root |
1.21 |
$txn->{finished}->recv; |
1513 |
root |
1.4 |
return $txn->{result}; |
1514 |
root |
1.3 |
|
1515 |
|
|
The actual code goes further and collects all errors ("die"s, |
1516 |
root |
1.22 |
exceptions) that occurred during request processing. The "result" method |
1517 |
root |
1.16 |
detects whether an exception as thrown (it is stored inside the $txn |
1518 |
root |
1.3 |
object) and just throws the exception, which means connection errors and |
1519 |
root |
1.60 |
other problems get reported to the code that tries to use the result, |
1520 |
root |
1.3 |
not in a random callback. |
1521 |
|
|
|
1522 |
|
|
All of this enables the following usage styles: |
1523 |
|
|
|
1524 |
|
|
1. Blocking: |
1525 |
|
|
|
1526 |
|
|
my $data = $fcp->client_get ($url); |
1527 |
|
|
|
1528 |
root |
1.15 |
2. Blocking, but running in parallel: |
1529 |
root |
1.3 |
|
1530 |
|
|
my @datas = map $_->result, |
1531 |
|
|
map $fcp->txn_client_get ($_), |
1532 |
|
|
@urls; |
1533 |
|
|
|
1534 |
|
|
Both blocking examples work without the module user having to know |
1535 |
|
|
anything about events. |
1536 |
|
|
|
1537 |
root |
1.15 |
3a. Event-based in a main program, using any supported event module: |
1538 |
root |
1.3 |
|
1539 |
root |
1.15 |
use EV; |
1540 |
root |
1.3 |
|
1541 |
|
|
$fcp->txn_client_get ($url)->cb (sub { |
1542 |
|
|
my $txn = shift; |
1543 |
|
|
my $data = $txn->result; |
1544 |
|
|
... |
1545 |
|
|
}); |
1546 |
|
|
|
1547 |
root |
1.15 |
EV::loop; |
1548 |
root |
1.3 |
|
1549 |
|
|
3b. The module user could use AnyEvent, too: |
1550 |
|
|
|
1551 |
|
|
use AnyEvent; |
1552 |
|
|
|
1553 |
|
|
my $quit = AnyEvent->condvar; |
1554 |
|
|
|
1555 |
|
|
$fcp->txn_client_get ($url)->cb (sub { |
1556 |
|
|
... |
1557 |
root |
1.21 |
$quit->send; |
1558 |
root |
1.3 |
}); |
1559 |
|
|
|
1560 |
root |
1.21 |
$quit->recv; |
1561 |
root |
1.3 |
|
1562 |
root |
1.19 |
BENCHMARKS |
1563 |
|
|
To give you an idea of the performance and overheads that AnyEvent adds |
1564 |
|
|
over the event loops themselves and to give you an impression of the |
1565 |
|
|
speed of various event loops I prepared some benchmarks. |
1566 |
|
|
|
1567 |
|
|
BENCHMARKING ANYEVENT OVERHEAD |
1568 |
|
|
Here is a benchmark of various supported event models used natively and |
1569 |
root |
1.22 |
through AnyEvent. The benchmark creates a lot of timers (with a zero |
1570 |
root |
1.19 |
timeout) and I/O watchers (watching STDOUT, a pty, to become writable, |
1571 |
|
|
which it is), lets them fire exactly once and destroys them again. |
1572 |
|
|
|
1573 |
|
|
Source code for this benchmark is found as eg/bench in the AnyEvent |
1574 |
root |
1.51 |
distribution. It uses the AE interface, which makes a real difference |
1575 |
|
|
for the EV and Perl backends only. |
1576 |
root |
1.19 |
|
1577 |
|
|
Explanation of the columns |
1578 |
|
|
*watcher* is the number of event watchers created/destroyed. Since |
1579 |
|
|
different event models feature vastly different performances, each event |
1580 |
|
|
loop was given a number of watchers so that overall runtime is |
1581 |
|
|
acceptable and similar between tested event loop (and keep them from |
1582 |
|
|
crashing): Glib would probably take thousands of years if asked to |
1583 |
|
|
process the same number of watchers as EV in this benchmark. |
1584 |
|
|
|
1585 |
|
|
*bytes* is the number of bytes (as measured by the resident set size, |
1586 |
|
|
RSS) consumed by each watcher. This method of measuring captures both C |
1587 |
|
|
and Perl-based overheads. |
1588 |
|
|
|
1589 |
|
|
*create* is the time, in microseconds (millionths of seconds), that it |
1590 |
|
|
takes to create a single watcher. The callback is a closure shared |
1591 |
|
|
between all watchers, to avoid adding memory overhead. That means |
1592 |
|
|
closure creation and memory usage is not included in the figures. |
1593 |
|
|
|
1594 |
|
|
*invoke* is the time, in microseconds, used to invoke a simple callback. |
1595 |
|
|
The callback simply counts down a Perl variable and after it was invoked |
1596 |
root |
1.21 |
"watcher" times, it would "->send" a condvar once to signal the end of |
1597 |
|
|
this phase. |
1598 |
root |
1.19 |
|
1599 |
|
|
*destroy* is the time, in microseconds, that it takes to destroy a |
1600 |
|
|
single watcher. |
1601 |
|
|
|
1602 |
|
|
Results |
1603 |
|
|
name watchers bytes create invoke destroy comment |
1604 |
root |
1.51 |
EV/EV 100000 223 0.47 0.43 0.27 EV native interface |
1605 |
|
|
EV/Any 100000 223 0.48 0.42 0.26 EV + AnyEvent watchers |
1606 |
|
|
Coro::EV/Any 100000 223 0.47 0.42 0.26 coroutines + Coro::Signal |
1607 |
|
|
Perl/Any 100000 431 2.70 0.74 0.92 pure perl implementation |
1608 |
|
|
Event/Event 16000 516 31.16 31.84 0.82 Event native interface |
1609 |
|
|
Event/Any 16000 1203 42.61 34.79 1.80 Event + AnyEvent watchers |
1610 |
|
|
IOAsync/Any 16000 1911 41.92 27.45 16.81 via IO::Async::Loop::IO_Poll |
1611 |
|
|
IOAsync/Any 16000 1726 40.69 26.37 15.25 via IO::Async::Loop::Epoll |
1612 |
|
|
Glib/Any 16000 1118 89.00 12.57 51.17 quadratic behaviour |
1613 |
|
|
Tk/Any 2000 1346 20.96 10.75 8.00 SEGV with >> 2000 watchers |
1614 |
|
|
POE/Any 2000 6951 108.97 795.32 14.24 via POE::Loop::Event |
1615 |
|
|
POE/Any 2000 6648 94.79 774.40 575.51 via POE::Loop::Select |
1616 |
root |
1.19 |
|
1617 |
|
|
Discussion |
1618 |
|
|
The benchmark does *not* measure scalability of the event loop very |
1619 |
|
|
well. For example, a select-based event loop (such as the pure perl one) |
1620 |
|
|
can never compete with an event loop that uses epoll when the number of |
1621 |
|
|
file descriptors grows high. In this benchmark, all events become ready |
1622 |
|
|
at the same time, so select/poll-based implementations get an unnatural |
1623 |
|
|
speed boost. |
1624 |
|
|
|
1625 |
|
|
Also, note that the number of watchers usually has a nonlinear effect on |
1626 |
|
|
overall speed, that is, creating twice as many watchers doesn't take |
1627 |
|
|
twice the time - usually it takes longer. This puts event loops tested |
1628 |
|
|
with a higher number of watchers at a disadvantage. |
1629 |
|
|
|
1630 |
|
|
To put the range of results into perspective, consider that on the |
1631 |
|
|
benchmark machine, handling an event takes roughly 1600 CPU cycles with |
1632 |
|
|
EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 |
1633 |
|
|
CPU cycles with POE. |
1634 |
|
|
|
1635 |
|
|
"EV" is the sole leader regarding speed and memory use, which are both |
1636 |
root |
1.51 |
maximal/minimal, respectively. When using the AE API there is zero |
1637 |
|
|
overhead (when going through the AnyEvent API create is about 5-6 times |
1638 |
|
|
slower, with other times being equal, so still uses far less memory than |
1639 |
|
|
any other event loop and is still faster than Event natively). |
1640 |
root |
1.19 |
|
1641 |
|
|
The pure perl implementation is hit in a few sweet spots (both the |
1642 |
|
|
constant timeout and the use of a single fd hit optimisations in the |
1643 |
|
|
perl interpreter and the backend itself). Nevertheless this shows that |
1644 |
|
|
it adds very little overhead in itself. Like any select-based backend |
1645 |
|
|
its performance becomes really bad with lots of file descriptors (and |
1646 |
|
|
few of them active), of course, but this was not subject of this |
1647 |
|
|
benchmark. |
1648 |
|
|
|
1649 |
|
|
The "Event" module has a relatively high setup and callback invocation |
1650 |
|
|
cost, but overall scores in on the third place. |
1651 |
|
|
|
1652 |
root |
1.41 |
"IO::Async" performs admirably well, about on par with "Event", even |
1653 |
|
|
when using its pure perl backend. |
1654 |
|
|
|
1655 |
root |
1.19 |
"Glib"'s memory usage is quite a bit higher, but it features a faster |
1656 |
|
|
callback invocation and overall ends up in the same class as "Event". |
1657 |
|
|
However, Glib scales extremely badly, doubling the number of watchers |
1658 |
|
|
increases the processing time by more than a factor of four, making it |
1659 |
|
|
completely unusable when using larger numbers of watchers (note that |
1660 |
|
|
only a single file descriptor was used in the benchmark, so |
1661 |
|
|
inefficiencies of "poll" do not account for this). |
1662 |
|
|
|
1663 |
|
|
The "Tk" adaptor works relatively well. The fact that it crashes with |
1664 |
|
|
more than 2000 watchers is a big setback, however, as correctness takes |
1665 |
|
|
precedence over speed. Nevertheless, its performance is surprising, as |
1666 |
|
|
the file descriptor is dup()ed for each watcher. This shows that the |
1667 |
|
|
dup() employed by some adaptors is not a big performance issue (it does |
1668 |
|
|
incur a hidden memory cost inside the kernel which is not reflected in |
1669 |
|
|
the figures above). |
1670 |
|
|
|
1671 |
|
|
"POE", regardless of underlying event loop (whether using its pure perl |
1672 |
|
|
select-based backend or the Event module, the POE-EV backend couldn't be |
1673 |
|
|
tested because it wasn't working) shows abysmal performance and memory |
1674 |
root |
1.20 |
usage with AnyEvent: Watchers use almost 30 times as much memory as EV |
1675 |
|
|
watchers, and 10 times as much memory as Event (the high memory |
1676 |
|
|
requirements are caused by requiring a session for each watcher). |
1677 |
|
|
Watcher invocation speed is almost 900 times slower than with AnyEvent's |
1678 |
|
|
pure perl implementation. |
1679 |
|
|
|
1680 |
|
|
The design of the POE adaptor class in AnyEvent can not really account |
1681 |
|
|
for the performance issues, though, as session creation overhead is |
1682 |
|
|
small compared to execution of the state machine, which is coded pretty |
1683 |
|
|
optimally within AnyEvent::Impl::POE (and while everybody agrees that |
1684 |
|
|
using multiple sessions is not a good approach, especially regarding |
1685 |
|
|
memory usage, even the author of POE could not come up with a faster |
1686 |
|
|
design). |
1687 |
root |
1.19 |
|
1688 |
|
|
Summary |
1689 |
|
|
* Using EV through AnyEvent is faster than any other event loop (even |
1690 |
|
|
when used without AnyEvent), but most event loops have acceptable |
1691 |
|
|
performance with or without AnyEvent. |
1692 |
|
|
|
1693 |
|
|
* The overhead AnyEvent adds is usually much smaller than the overhead |
1694 |
|
|
of the actual event loop, only with extremely fast event loops such |
1695 |
root |
1.66 |
as EV does AnyEvent add significant overhead. |
1696 |
root |
1.19 |
|
1697 |
|
|
* You should avoid POE like the plague if you want performance or |
1698 |
|
|
reasonable memory usage. |
1699 |
|
|
|
1700 |
|
|
BENCHMARKING THE LARGE SERVER CASE |
1701 |
root |
1.22 |
This benchmark actually benchmarks the event loop itself. It works by |
1702 |
|
|
creating a number of "servers": each server consists of a socket pair, a |
1703 |
root |
1.19 |
timeout watcher that gets reset on activity (but never fires), and an |
1704 |
|
|
I/O watcher waiting for input on one side of the socket. Each time the |
1705 |
|
|
socket watcher reads a byte it will write that byte to a random other |
1706 |
|
|
"server". |
1707 |
|
|
|
1708 |
|
|
The effect is that there will be a lot of I/O watchers, only part of |
1709 |
|
|
which are active at any one point (so there is a constant number of |
1710 |
root |
1.22 |
active fds for each loop iteration, but which fds these are is random). |
1711 |
root |
1.19 |
The timeout is reset each time something is read because that reflects |
1712 |
|
|
how most timeouts work (and puts extra pressure on the event loops). |
1713 |
|
|
|
1714 |
root |
1.22 |
In this benchmark, we use 10000 socket pairs (20000 sockets), of which |
1715 |
root |
1.19 |
100 (1%) are active. This mirrors the activity of large servers with |
1716 |
|
|
many connections, most of which are idle at any one point in time. |
1717 |
|
|
|
1718 |
|
|
Source code for this benchmark is found as eg/bench2 in the AnyEvent |
1719 |
root |
1.51 |
distribution. It uses the AE interface, which makes a real difference |
1720 |
|
|
for the EV and Perl backends only. |
1721 |
root |
1.19 |
|
1722 |
|
|
Explanation of the columns |
1723 |
|
|
*sockets* is the number of sockets, and twice the number of "servers" |
1724 |
|
|
(as each server has a read and write socket end). |
1725 |
|
|
|
1726 |
root |
1.22 |
*create* is the time it takes to create a socket pair (which is |
1727 |
root |
1.19 |
nontrivial) and two watchers: an I/O watcher and a timeout watcher. |
1728 |
|
|
|
1729 |
|
|
*request*, the most important value, is the time it takes to handle a |
1730 |
|
|
single "request", that is, reading the token from the pipe and |
1731 |
|
|
forwarding it to another server. This includes deleting the old timeout |
1732 |
|
|
and creating a new one that moves the timeout into the future. |
1733 |
|
|
|
1734 |
|
|
Results |
1735 |
root |
1.41 |
name sockets create request |
1736 |
root |
1.51 |
EV 20000 62.66 7.99 |
1737 |
|
|
Perl 20000 68.32 32.64 |
1738 |
|
|
IOAsync 20000 174.06 101.15 epoll |
1739 |
|
|
IOAsync 20000 174.67 610.84 poll |
1740 |
|
|
Event 20000 202.69 242.91 |
1741 |
|
|
Glib 20000 557.01 1689.52 |
1742 |
|
|
POE 20000 341.54 12086.32 uses POE::Loop::Event |
1743 |
root |
1.19 |
|
1744 |
|
|
Discussion |
1745 |
|
|
This benchmark *does* measure scalability and overall performance of the |
1746 |
|
|
particular event loop. |
1747 |
|
|
|
1748 |
|
|
EV is again fastest. Since it is using epoll on my system, the setup |
1749 |
|
|
time is relatively high, though. |
1750 |
|
|
|
1751 |
|
|
Perl surprisingly comes second. It is much faster than the C-based event |
1752 |
|
|
loops Event and Glib. |
1753 |
|
|
|
1754 |
root |
1.41 |
IO::Async performs very well when using its epoll backend, and still |
1755 |
|
|
quite good compared to Glib when using its pure perl backend. |
1756 |
|
|
|
1757 |
root |
1.19 |
Event suffers from high setup time as well (look at its code and you |
1758 |
|
|
will understand why). Callback invocation also has a high overhead |
1759 |
|
|
compared to the "$_->() for .."-style loop that the Perl event loop |
1760 |
|
|
uses. Event uses select or poll in basically all documented |
1761 |
|
|
configurations. |
1762 |
|
|
|
1763 |
|
|
Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It |
1764 |
|
|
clearly fails to perform with many filehandles or in busy servers. |
1765 |
|
|
|
1766 |
|
|
POE is still completely out of the picture, taking over 1000 times as |
1767 |
|
|
long as EV, and over 100 times as long as the Perl implementation, even |
1768 |
|
|
though it uses a C-based event loop in this case. |
1769 |
|
|
|
1770 |
|
|
Summary |
1771 |
root |
1.20 |
* The pure perl implementation performs extremely well. |
1772 |
root |
1.19 |
|
1773 |
|
|
* Avoid Glib or POE in large projects where performance matters. |
1774 |
|
|
|
1775 |
|
|
BENCHMARKING SMALL SERVERS |
1776 |
|
|
While event loops should scale (and select-based ones do not...) even to |
1777 |
|
|
large servers, most programs we (or I :) actually write have only a few |
1778 |
|
|
I/O watchers. |
1779 |
|
|
|
1780 |
|
|
In this benchmark, I use the same benchmark program as in the large |
1781 |
|
|
server case, but it uses only eight "servers", of which three are active |
1782 |
|
|
at any one time. This should reflect performance for a small server |
1783 |
|
|
relatively well. |
1784 |
|
|
|
1785 |
|
|
The columns are identical to the previous table. |
1786 |
|
|
|
1787 |
|
|
Results |
1788 |
|
|
name sockets create request |
1789 |
|
|
EV 16 20.00 6.54 |
1790 |
|
|
Perl 16 25.75 12.62 |
1791 |
|
|
Event 16 81.27 35.86 |
1792 |
|
|
Glib 16 32.63 15.48 |
1793 |
|
|
POE 16 261.87 276.28 uses POE::Loop::Event |
1794 |
|
|
|
1795 |
|
|
Discussion |
1796 |
|
|
The benchmark tries to test the performance of a typical small server. |
1797 |
|
|
While knowing how various event loops perform is interesting, keep in |
1798 |
|
|
mind that their overhead in this case is usually not as important, due |
1799 |
|
|
to the small absolute number of watchers (that is, you need efficiency |
1800 |
|
|
and speed most when you have lots of watchers, not when you only have a |
1801 |
|
|
few of them). |
1802 |
|
|
|
1803 |
|
|
EV is again fastest. |
1804 |
|
|
|
1805 |
root |
1.22 |
Perl again comes second. It is noticeably faster than the C-based event |
1806 |
root |
1.19 |
loops Event and Glib, although the difference is too small to really |
1807 |
|
|
matter. |
1808 |
|
|
|
1809 |
|
|
POE also performs much better in this case, but is is still far behind |
1810 |
|
|
the others. |
1811 |
|
|
|
1812 |
|
|
Summary |
1813 |
|
|
* C-based event loops perform very well with small number of watchers, |
1814 |
|
|
as the management overhead dominates. |
1815 |
|
|
|
1816 |
root |
1.40 |
THE IO::Lambda BENCHMARK |
1817 |
|
|
Recently I was told about the benchmark in the IO::Lambda manpage, which |
1818 |
|
|
could be misinterpreted to make AnyEvent look bad. In fact, the |
1819 |
|
|
benchmark simply compares IO::Lambda with POE, and IO::Lambda looks |
1820 |
|
|
better (which shouldn't come as a surprise to anybody). As such, the |
1821 |
root |
1.41 |
benchmark is fine, and mostly shows that the AnyEvent backend from |
1822 |
|
|
IO::Lambda isn't very optimal. But how would AnyEvent compare when used |
1823 |
|
|
without the extra baggage? To explore this, I wrote the equivalent |
1824 |
|
|
benchmark for AnyEvent. |
1825 |
root |
1.40 |
|
1826 |
|
|
The benchmark itself creates an echo-server, and then, for 500 times, |
1827 |
|
|
connects to the echo server, sends a line, waits for the reply, and then |
1828 |
|
|
creates the next connection. This is a rather bad benchmark, as it |
1829 |
root |
1.41 |
doesn't test the efficiency of the framework or much non-blocking I/O, |
1830 |
|
|
but it is a benchmark nevertheless. |
1831 |
root |
1.40 |
|
1832 |
|
|
name runtime |
1833 |
|
|
Lambda/select 0.330 sec |
1834 |
|
|
+ optimized 0.122 sec |
1835 |
|
|
Lambda/AnyEvent 0.327 sec |
1836 |
|
|
+ optimized 0.138 sec |
1837 |
|
|
Raw sockets/select 0.077 sec |
1838 |
|
|
POE/select, components 0.662 sec |
1839 |
|
|
POE/select, raw sockets 0.226 sec |
1840 |
|
|
POE/select, optimized 0.404 sec |
1841 |
|
|
|
1842 |
|
|
AnyEvent/select/nb 0.085 sec |
1843 |
|
|
AnyEvent/EV/nb 0.068 sec |
1844 |
|
|
+state machine 0.134 sec |
1845 |
|
|
|
1846 |
root |
1.41 |
The benchmark is also a bit unfair (my fault): the IO::Lambda/POE |
1847 |
root |
1.40 |
benchmarks actually make blocking connects and use 100% blocking I/O, |
1848 |
|
|
defeating the purpose of an event-based solution. All of the newly |
1849 |
|
|
written AnyEvent benchmarks use 100% non-blocking connects (using |
1850 |
|
|
AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS |
1851 |
root |
1.41 |
resolver), so AnyEvent is at a disadvantage here, as non-blocking |
1852 |
root |
1.40 |
connects generally require a lot more bookkeeping and event handling |
1853 |
|
|
than blocking connects (which involve a single syscall only). |
1854 |
|
|
|
1855 |
|
|
The last AnyEvent benchmark additionally uses AnyEvent::Handle, which |
1856 |
root |
1.41 |
offers similar expressive power as POE and IO::Lambda, using |
1857 |
|
|
conventional Perl syntax. This means that both the echo server and the |
1858 |
|
|
client are 100% non-blocking, further placing it at a disadvantage. |
1859 |
|
|
|
1860 |
|
|
As you can see, the AnyEvent + EV combination even beats the |
1861 |
|
|
hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl |
1862 |
|
|
backend easily beats IO::Lambda and POE. |
1863 |
root |
1.40 |
|
1864 |
|
|
And even the 100% non-blocking version written using the high-level (and |
1865 |
root |
1.54 |
slow :) AnyEvent::Handle abstraction beats both POE and IO::Lambda |
1866 |
|
|
higher level ("unoptimised") abstractions by a large margin, even though |
1867 |
|
|
it does all of DNS, tcp-connect and socket I/O in a non-blocking way. |
1868 |
root |
1.41 |
|
1869 |
|
|
The two AnyEvent benchmarks programs can be found as eg/ae0.pl and |
1870 |
|
|
eg/ae2.pl in the AnyEvent distribution, the remaining benchmarks are |
1871 |
root |
1.54 |
part of the IO::Lambda distribution and were used without any changes. |
1872 |
root |
1.40 |
|
1873 |
root |
1.32 |
SIGNALS |
1874 |
|
|
AnyEvent currently installs handlers for these signals: |
1875 |
|
|
|
1876 |
|
|
SIGCHLD |
1877 |
|
|
A handler for "SIGCHLD" is installed by AnyEvent's child watcher |
1878 |
|
|
emulation for event loops that do not support them natively. Also, |
1879 |
|
|
some event loops install a similar handler. |
1880 |
|
|
|
1881 |
root |
1.44 |
Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE, |
1882 |
|
|
then AnyEvent will reset it to default, to avoid losing child exit |
1883 |
|
|
statuses. |
1884 |
root |
1.41 |
|
1885 |
root |
1.32 |
SIGPIPE |
1886 |
|
|
A no-op handler is installed for "SIGPIPE" when $SIG{PIPE} is |
1887 |
|
|
"undef" when AnyEvent gets loaded. |
1888 |
|
|
|
1889 |
|
|
The rationale for this is that AnyEvent users usually do not really |
1890 |
|
|
depend on SIGPIPE delivery (which is purely an optimisation for |
1891 |
|
|
shell use, or badly-written programs), but "SIGPIPE" can cause |
1892 |
|
|
spurious and rare program exits as a lot of people do not expect |
1893 |
|
|
"SIGPIPE" when writing to some random socket. |
1894 |
|
|
|
1895 |
|
|
The rationale for installing a no-op handler as opposed to ignoring |
1896 |
|
|
it is that this way, the handler will be restored to defaults on |
1897 |
|
|
exec. |
1898 |
|
|
|
1899 |
|
|
Feel free to install your own handler, or reset it to defaults. |
1900 |
|
|
|
1901 |
root |
1.46 |
RECOMMENDED/OPTIONAL MODULES |
1902 |
|
|
One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and |
1903 |
root |
1.63 |
its built-in modules) are required to use it. |
1904 |
root |
1.46 |
|
1905 |
|
|
That does not mean that AnyEvent won't take advantage of some additional |
1906 |
|
|
modules if they are installed. |
1907 |
|
|
|
1908 |
root |
1.57 |
This section explains which additional modules will be used, and how |
1909 |
|
|
they affect AnyEvent's operation. |
1910 |
root |
1.46 |
|
1911 |
|
|
Async::Interrupt |
1912 |
|
|
This slightly arcane module is used to implement fast signal |
1913 |
|
|
handling: To my knowledge, there is no way to do completely |
1914 |
|
|
race-free and quick signal handling in pure perl. To ensure that |
1915 |
|
|
signals still get delivered, AnyEvent will start an interval timer |
1916 |
root |
1.47 |
to wake up perl (and catch the signals) with some delay (default is |
1917 |
root |
1.46 |
10 seconds, look for $AnyEvent::MAX_SIGNAL_LATENCY). |
1918 |
|
|
|
1919 |
|
|
If this module is available, then it will be used to implement |
1920 |
|
|
signal catching, which means that signals will not be delayed, and |
1921 |
|
|
the event loop will not be interrupted regularly, which is more |
1922 |
root |
1.57 |
efficient (and good for battery life on laptops). |
1923 |
root |
1.46 |
|
1924 |
|
|
This affects not just the pure-perl event loop, but also other event |
1925 |
|
|
loops that have no signal handling on their own (e.g. Glib, Tk, Qt). |
1926 |
|
|
|
1927 |
root |
1.47 |
Some event loops (POE, Event, Event::Lib) offer signal watchers |
1928 |
|
|
natively, and either employ their own workarounds (POE) or use |
1929 |
|
|
AnyEvent's workaround (using $AnyEvent::MAX_SIGNAL_LATENCY). |
1930 |
|
|
Installing Async::Interrupt does nothing for those backends. |
1931 |
|
|
|
1932 |
root |
1.46 |
EV This module isn't really "optional", as it is simply one of the |
1933 |
|
|
backend event loops that AnyEvent can use. However, it is simply the |
1934 |
|
|
best event loop available in terms of features, speed and stability: |
1935 |
|
|
It supports the AnyEvent API optimally, implements all the watcher |
1936 |
|
|
types in XS, does automatic timer adjustments even when no monotonic |
1937 |
|
|
clock is available, can take avdantage of advanced kernel interfaces |
1938 |
|
|
such as "epoll" and "kqueue", and is the fastest backend *by far*. |
1939 |
|
|
You can even embed Glib/Gtk2 in it (or vice versa, see EV::Glib and |
1940 |
|
|
Glib::EV). |
1941 |
|
|
|
1942 |
root |
1.60 |
If you only use backends that rely on another event loop (e.g. |
1943 |
|
|
"Tk"), then this module will do nothing for you. |
1944 |
|
|
|
1945 |
root |
1.46 |
Guard |
1946 |
|
|
The guard module, when used, will be used to implement |
1947 |
|
|
"AnyEvent::Util::guard". This speeds up guards considerably (and |
1948 |
|
|
uses a lot less memory), but otherwise doesn't affect guard |
1949 |
|
|
operation much. It is purely used for performance. |
1950 |
|
|
|
1951 |
|
|
JSON and JSON::XS |
1952 |
root |
1.55 |
One of these modules is required when you want to read or write JSON |
1953 |
root |
1.60 |
data via AnyEvent::Handle. JSON is also written in pure-perl, but |
1954 |
|
|
can take advantage of the ultra-high-speed JSON::XS module when it |
1955 |
|
|
is installed. |
1956 |
root |
1.46 |
|
1957 |
|
|
Net::SSLeay |
1958 |
|
|
Implementing TLS/SSL in Perl is certainly interesting, but not very |
1959 |
|
|
worthwhile: If this module is installed, then AnyEvent::Handle (with |
1960 |
|
|
the help of AnyEvent::TLS), gains the ability to do TLS/SSL. |
1961 |
|
|
|
1962 |
|
|
Time::HiRes |
1963 |
|
|
This module is part of perl since release 5.008. It will be used |
1964 |
root |
1.63 |
when the chosen event library does not come with a timing source of |
1965 |
root |
1.65 |
its own. The pure-perl event loop (AnyEvent::Loop) will additionally |
1966 |
|
|
load it to try to use a monotonic clock for timing stability. |
1967 |
root |
1.46 |
|
1968 |
root |
1.18 |
FORK |
1969 |
|
|
Most event libraries are not fork-safe. The ones who are usually are |
1970 |
root |
1.59 |
because they rely on inefficient but fork-safe "select" or "poll" calls |
1971 |
|
|
- higher performance APIs such as BSD's kqueue or the dreaded Linux |
1972 |
|
|
epoll are usually badly thought-out hacks that are incompatible with |
1973 |
|
|
fork in one way or another. Only EV is fully fork-aware and ensures that |
1974 |
|
|
you continue event-processing in both parent and child (or both, if you |
1975 |
|
|
know what you are doing). |
1976 |
root |
1.18 |
|
1977 |
root |
1.57 |
This means that, in general, you cannot fork and do event processing in |
1978 |
root |
1.59 |
the child if the event library was initialised before the fork (which |
1979 |
|
|
usually happens when the first AnyEvent watcher is created, or the |
1980 |
|
|
library is loaded). |
1981 |
root |
1.57 |
|
1982 |
root |
1.18 |
If you have to fork, you must either do so *before* creating your first |
1983 |
root |
1.46 |
watcher OR you must not use AnyEvent at all in the child OR you must do |
1984 |
|
|
something completely out of the scope of AnyEvent. |
1985 |
root |
1.18 |
|
1986 |
root |
1.57 |
The problem of doing event processing in the parent *and* the child is |
1987 |
|
|
much more complicated: even for backends that *are* fork-aware or |
1988 |
|
|
fork-safe, their behaviour is not usually what you want: fork clones all |
1989 |
|
|
watchers, that means all timers, I/O watchers etc. are active in both |
1990 |
root |
1.59 |
parent and child, which is almost never what you want. USing "exec" to |
1991 |
|
|
start worker children from some kind of manage rprocess is usually |
1992 |
|
|
preferred, because it is much easier and cleaner, at the expense of |
1993 |
|
|
having to have another binary. |
1994 |
root |
1.57 |
|
1995 |
root |
1.18 |
SECURITY CONSIDERATIONS |
1996 |
|
|
AnyEvent can be forced to load any event model via |
1997 |
|
|
$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used |
1998 |
|
|
to execute arbitrary code or directly gain access, it can easily be used |
1999 |
|
|
to make the program hang or malfunction in subtle ways, as AnyEvent |
2000 |
|
|
watchers will not be active when the program uses a different event |
2001 |
|
|
model than specified in the variable. |
2002 |
|
|
|
2003 |
|
|
You can make AnyEvent completely ignore this variable by deleting it |
2004 |
|
|
before the first watcher gets created, e.g. with a "BEGIN" block: |
2005 |
|
|
|
2006 |
root |
1.25 |
BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} } |
2007 |
root |
1.62 |
|
2008 |
|
|
use AnyEvent; |
2009 |
root |
1.18 |
|
2010 |
root |
1.20 |
Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can |
2011 |
|
|
be used to probe what backend is used and gain other information (which |
2012 |
root |
1.28 |
is probably even less useful to an attacker than PERL_ANYEVENT_MODEL), |
2013 |
root |
1.40 |
and $ENV{PERL_ANYEVENT_STRICT}. |
2014 |
root |
1.20 |
|
2015 |
root |
1.41 |
Note that AnyEvent will remove *all* environment variables starting with |
2016 |
|
|
"PERL_ANYEVENT_" from %ENV when it is loaded while taint mode is |
2017 |
|
|
enabled. |
2018 |
|
|
|
2019 |
root |
1.26 |
BUGS |
2020 |
|
|
Perl 5.8 has numerous memleaks that sometimes hit this module and are |
2021 |
|
|
hard to work around. If you suffer from memleaks, first upgrade to Perl |
2022 |
|
|
5.10 and check wether the leaks still show up. (Perl 5.10.0 has other |
2023 |
root |
1.36 |
annoying memleaks, such as leaking on "map" and "grep" but it is usually |
2024 |
root |
1.26 |
not as pronounced). |
2025 |
|
|
|
2026 |
root |
1.2 |
SEE ALSO |
2027 |
root |
1.63 |
Tutorial/Introduction: AnyEvent::Intro. |
2028 |
|
|
|
2029 |
|
|
FAQ: AnyEvent::FAQ. |
2030 |
|
|
|
2031 |
root |
1.66 |
Utility functions: AnyEvent::Util (misc. grab-bag), AnyEvent::Log |
2032 |
|
|
(simply logging). |
2033 |
|
|
|
2034 |
|
|
Development/Debugging: AnyEvent::Strict (stricter checking), |
2035 |
|
|
AnyEvent::Debug (interactive shell, watcher tracing). |
2036 |
root |
1.22 |
|
2037 |
root |
1.66 |
Supported event modules: AnyEvent::Loop, EV, EV::Glib, Glib::EV, Event, |
2038 |
|
|
Glib::Event, Glib, Tk, Event::Lib, Qt, POE, FLTK. |
2039 |
root |
1.20 |
|
2040 |
|
|
Implementations: AnyEvent::Impl::EV, AnyEvent::Impl::Event, |
2041 |
|
|
AnyEvent::Impl::Glib, AnyEvent::Impl::Tk, AnyEvent::Impl::Perl, |
2042 |
root |
1.43 |
AnyEvent::Impl::EventLib, AnyEvent::Impl::Qt, AnyEvent::Impl::POE, |
2043 |
root |
1.66 |
AnyEvent::Impl::IOAsync, Anyevent::Impl::Irssi, AnyEvent::Impl::FLTK. |
2044 |
root |
1.3 |
|
2045 |
root |
1.66 |
Non-blocking handles, pipes, stream sockets, TCP clients and servers: |
2046 |
root |
1.43 |
AnyEvent::Handle, AnyEvent::Socket, AnyEvent::TLS. |
2047 |
root |
1.22 |
|
2048 |
|
|
Asynchronous DNS: AnyEvent::DNS. |
2049 |
|
|
|
2050 |
root |
1.63 |
Thread support: Coro, Coro::AnyEvent, Coro::EV, Coro::Event. |
2051 |
root |
1.3 |
|
2052 |
root |
1.63 |
Nontrivial usage examples: AnyEvent::GPSD, AnyEvent::IRC, |
2053 |
root |
1.43 |
AnyEvent::HTTP. |
2054 |
root |
1.2 |
|
2055 |
root |
1.17 |
AUTHOR |
2056 |
root |
1.25 |
Marc Lehmann <schmorp@schmorp.de> |
2057 |
|
|
http://home.schmorp.de/ |
2058 |
root |
1.2 |
|